Protective measures and devices for parking and / or loading areas

Abstract

One method for detecting battery fires in electric or hybrid vehicles involves capturing thermal images of the vehicle's underside to distinguish between combustion engine and electric or hybrid vehicles and to track the thermal profile of high-temperature areas over time. Various operating modes enable, among other things, energy-efficient, regular monitoring and highly precise monitoring in situations with irregular thermal profiles.

Description

AREA OF INVENTION

[0001] The present invention relates to protective methods and devices for monitoring parking and / or charging points for vehicles. BACKGROUND

[0002] Modern cars, especially hybrid and fully electric vehicles, contain batteries that store ever-increasing amounts of electrical energy for propulsion. Due to the steadily increasing energy density of these batteries, the potential for damage also increases if such a battery catches fire. A battery fire is very difficult to control and extinguish, particularly with most current lithium-based battery types.

[0003] If a battery catches fire despite all precautions, water is the best way to control the blaze, but it won't do much to extinguish the burning battery directly, as lithium floats on water and will even continue to burn upon contact. While it's possible to extinguish a lithium battery fire by completely submerging the car—and the battery—in water, this is often not a realistic option in multi-story parking garages. However, water can still help by protecting surrounding objects and surfaces from the fire.

[0004] Special fire extinguishers are required to extinguish burning lithium batteries. These so-called AVD fire extinguishers are filled with gel or silicate. The ingredients cool or solidify instantly, simultaneously depriving the fire of oxygen. This is important because battery fires can reignite. Toxic fumes, which are also produced during battery fires, can be contained if the gel or silicate forms a closed barrier.

[0005] Smothering battery fires with blankets, sand, or other extinguishing agents does not yield satisfactory results, as the thermal energy from the chemical processes must be dissipated; that is, the chemical processes must be cooled to effectively extinguish the fire. Since the oxygen required for combustion is already chemically bound within the cells, simply covering or smothering the source of the fire is insufficient.

[0006] If a battery fire occurs on an open street, it will remain a localized event and most likely only damage the vehicle directly affected. The likelihood of more extensive damage is much higher if a battery fire occurs in a public or private garage with potentially hundreds of cars parked side-by-side, or in areas with flammable materials nearby.

[0007] Most battery fires are caused by mechanical damage to the battery cells. In most vehicles, the relatively heavy batteries are located near the underbody to keep the center of gravity as low as possible. This makes the batteries vulnerable to damage unless protected by appropriate measures. Despite all protective measures, stones can be thrown against the underbody of the vehicle where the battery pack is located while driving, or the vehicle's undercarriage can twist when driving over extremely uneven terrain, which can damage the battery pack located under the vehicle.Damage to the battery or its auxiliary systems, such as cooling and heating, does not necessarily cause an immediate fire; rather, the condition of the damaged battery can deteriorate slowly until it eventually ignites when the necessary conditions are met. Other causes of battery fires include thermal instability due to overcharging or malfunctions of the battery management system.

[0008] Like any other primary cell or battery, a rechargeable battery has two volumes, one on the anode side and one on the cathode side. The charge carriers—in this case, the lithium ions—flow back and forth between these two volumes, depending on whether the battery is being discharged or charged. To ensure that this happens in a controlled manner, a separator, a semi-permeable partition, is positioned between the volumes. Semi-permeable means that only the charge carriers in the battery—that is, the lithium ions—can pass through the separator, and then only slowly and in controlled amounts.

[0009] If the separator is damaged—whether by excessive heat, mechanical penetration, or simply a manufacturing defect—this partition becomes more permeable to the lithium ions than it should be. In other words, a short circuit occurs. This causes the lithium ions to migrate very rapidly between the two volumes of the battery, and chemical reactions proceed very quickly, releasing a large amount of energy in a short time that cannot be dissipated from the battery quickly enough. This leads to a chain reaction: First, the hotter the battery becomes, the more likely the separator is to fail; and second, the faster the reactions occur, the higher the temperature rises. During a battery thermal runaway, temperatures exceeding 1.000 °C can occur, which ignite many other materials found in modern vehicles and eventually burn the entire vehicle.

[0010] Currently, parking lots are equipped with a range of fire detection and extinguishing systems, most of which are designed for "conventional" vehicle fires. These systems include smoke detectors, heat detectors, carbon monoxide detectors (CO detectors), thermal imaging cameras, and sprinkler systems.

[0011] Smoke detectors are the most common devices and detect the presence of smoke in the air, which can indicate a fire. They are frequently used in underground parking garages where ventilation is limited.

[0012] Heat detectors are used to detect a rapid rise in air temperature in the event of a fire. These detectors may be better suited for environments where dust and exhaust fumes could impair the effectiveness of smoke detectors.

[0013] CO detectors can serve as an early warning system for a developing fire, especially in enclosed parking garages where vehicles are frequently in motion, since CO is a byproduct of combustion.

[0014] Although sprinkler systems are not suitable for fire detection, they can control a fire and often extinguish it before the fire brigade arrives, especially in enclosed areas such as underground car parks.

[0015] Thermal imaging cameras can detect abnormal temperature fluctuations and are useful for monitoring large areas, such as open parking lots or multi-story parking garages. However, current installations only observe vehicles from the side or above, which limits their ability to detect fires originating on the underside of a vehicle. For example, CN 1 17 475 574 A discloses a fire detection system in which an infrared camera captures the chassis of a vehicle upon entry and which also includes a smoke detector for each parking space. KR 10 2 672 274 B1 discloses another fire detection system in which one camera monitors an area equipped with charging stations for electric vehicles, and another camera at each charging station monitors the vehicle parked there.

[0016] DE 10 2021 105 472 A1 discloses a device for charging an electrical energy storage device located in a vehicle, which interrupts the charging if a thermal runaway of an electrical energy storage device is detected in a neighboring parked vehicle.

[0017] Finally, DE 10 2019 128 864 A1 discloses a parking management system which has a temperature sensor in the ground of each parking space and evaluates a temperature profile using an evaluation unit.

[0018] There is a need for a method and a system that enables improved detection of battery fires in electric or hybrid vehicles. DESCRIPTION OF THE INVENTION

[0019] This need is met by the method according to claim 1 and the system according to claim 9. Advantageous embodiments and further developments of the method and the system are specified in the respective dependent claims. A computer program product and a computer-readable medium or data carrier are specified in claims 11 and 12, respectively.

[0020] According to a first aspect of the present invention, a method for detecting battery fires in an electric or hybrid vehicle (EV) in a parking space comprises a first and a second operating mode. In the first operating mode, a detection is performed to determine whether the occupancy status of the parking space changes from empty to occupied, i.e., whether a vehicle is parked in the previously empty parking space. If so, a corresponding signal is generated. The first operating mode can further include detecting a continuing occupancy of the parking space, i.e., that the vehicle is still present, and detecting that the parking space is being vacated, i.e., that the previously parked vehicle has left the parking space. The signal is used as a trigger for entering the second operating mode.It should be noted that in this description the abbreviation EV is used for both fully electric vehicles and hybrid vehicles, unless explicitly stated otherwise or evident from the context.

[0021] The second operating mode, activated upon receiving a signal indicating that the parking space occupancy status has changed from empty to occupied, involves performing one or more first-type scans using a thermal sensor array over a first field of view to obtain a corresponding first-type thermal image. The thermal sensor array can include any type of sensor capable of detecting temperatures, including a camera capable of capturing a thermal image covering the first field of view, or a sensor capable of scanning the first field of view to obtain a complete thermal image. The first field of view can be characterized by a solid angle captured by the sensor and may be rectangular or otherwise shaped. The first field of view is conveniently directed upwards from the parking space floor, i.e., towards the underside of a vehicle parked in the parking space.The first-type thermal image is then analyzed to identify contiguous pixel clusters with high temperatures. Based on this first-type scan, the method determines whether the parked vehicle is an electric vehicle. This determination may involve capturing multiple thermal images over time and analyzing the distribution or changes in the high-temperature pixel clusters. Rapid cooling, i.e., a rapid decrease in the number of contiguous high-temperature pixels within the clusters, is strong evidence that the vehicle is equipped with an internal combustion engine, while a slower and / or gradual decrease in temperature within the contiguous cluster is strong evidence that the vehicle is an electric vehicle.Similarly, the distribution of contiguous high-temperature pixels can indicate the presence of an internal combustion engine, e.g., if a narrow cluster of high-temperature pixels extends over a longer section of the vehicle, this can be interpreted as an exhaust pipe.

[0022] If the parked vehicle is not an electric vehicle, i.e., if the determination step yields a negative result, the procedure can return to the first operating mode and remain in that mode until a new signal is received indicating that the parking space has been vacated and reoccupied.

[0023] If the parked vehicle is an electric vehicle, i.e., if the determination step yields a positive result, the system switches from the second operating mode to a third mode in which the method includes identifying an area within the first field of view of the thermal image for subsequent monitoring. In one or more embodiments, identifying an area within the first field of view of the obtained thermal image for subsequent monitoring includes identifying a contiguous cluster of high-temperature pixels in the obtained thermal image that exhibits the highest temperature.The subsequent monitoring involves performing second-type scans at initial time intervals over a second field of view encompassing the identified area, as long as the temperature of the identified area does not rise, as long as the temperature in the second field of view outside the identified area does not exceed the ambient temperature by a predetermined initial value, and as long as the parking space remains occupied. The term "time interval" can encompass time intervals of variable length, the length of which can be adjusted according to requirements and / or environmental conditions. The second field of view, which is smaller than the first, can, like the first, be characterized by a solid angle detected by the sensor and can have a rectangular or other shape.The ambient temperature can be provided by a suitable sensor, which may be part of a system implementing the method, or by a separate sensor located outside the system. It is also conceivable to determine the ambient temperature using the thermal sensor array of the system implementing the method, for example, by scans performed when a parking space is unoccupied. This latter embodiment, which may be particularly suitable or useful for parking spaces in buildings, takes advantage of the fact that the ambient temperature does not typically change rapidly over time.

[0024] The procedure further includes a fourth operating mode, which is invoked from the third operating mode if the temperature of the identified area rises or if the temperature in the second field of view outside the identified area exceeds the ambient temperature by more than the predetermined first value, as long as the parking space remains occupied. The fourth operating mode involves performing scans of a third type, covering at least the identified area, at second time intervals that are shorter than the first time intervals previously used when performing scans of the second type, as long as a temperature in any part of the area covered by the scans of the third type rises or does not fall within a predetermined first number of consecutive scans of the third type or within a predetermined first period, and as long as the parking space remains occupied.

[0025] In the fourth operating mode, if the temperature in any area covered by the third-type scans rises by a predetermined value or more within a predetermined second number of consecutive third-type scans or within a predetermined second time period, a first alarm signal is issued. The alarm signal can be sent to various recipients, including a facility manager or the fire department.

[0026] If, in the fourth operating mode, the temperature in any area covered by the third-type scans drops within the predetermined first number of consecutive third-type scans or within the predetermined first period, as long as the parking space remains occupied, the fourth operating mode is exited and the system returns to the third operating mode. This may also include issuing a signal to terminate a previously triggered alarm.

[0027] It should be noted that the first and second time intervals can be adaptively selected from ranges, and that the range from which the respective first time intervals are selected may overlap with the range from which the respective second time intervals are selected. Within the scope of this invention, the only condition regarding the respective first and second time intervals is that the second time intervals are shorter than the first time intervals.

[0028] The change in the occupancy status of the parking space from empty to occupied and / or the continued occupancy of the parking space, i.e., the continued—and uninterrupted—presence of the vehicle in the parking space, can be indicated by a signal received from an independent parking monitoring system or a corresponding sensor of a system implementing the procedure, or by the absence of such a signal. In particular, a signal indicating the continued occupancy of the parking space can additionally or alternatively be generated by a process that evaluates the first, second, third, and fourth type scans in the respective operating mode. In each of the scans, or rather in a sequence of successive scans, a sudden change, most likely a drop, in the temperature in the area identified for monitoring by a predetermined third value can indicate that the parking space has been vacated.

[0029] In one or more embodiments of the method for determining whether a parked vehicle is an electric vehicle, the method comprises identifying and evaluating the position of one or more contiguous high-temperature pixel clusters within the first field of view of the obtained thermal image. The evaluation may include taking into account the fact that electric vehicle batteries, which are warmer than most other areas of the underside of the electric vehicle, are typically located in a central part of the electric vehicle's chassis, whereas internal combustion engines, which are warmer than most other areas of the underside of internal combustion engine vehicles, are typically located in the front region.Additionally, to increase the reliability of the detection, or simply as an alternative solution, the temperature change of adjacent pixel clusters can be evaluated over a number of consecutive scans of the first type or over a predetermined third period. The temperature profiles of cooling combustion engines and EV batteries differ considerably, for example, in their initial temperatures after use, which can be exploited for this purpose. As mentioned above, the shape of contiguous high-temperature pixel clusters can also be used to determine the presence of an electric vehicle. Hot exhaust pipes of combustion engines are typically narrow and elongated, while batteries usually cover a wider, contiguous area. As a further addition or alternative, the detection can include receiving a signal indicating that a charging device connected to the parking space is activated.This latter alternative allows for the identification of hybrid vehicles that were driven using their combustion engine and are now being charged while parked. Since battery fires can occur during charging, this further increases the reliability of the procedure and the system that implements it.

[0030] In one or more embodiments of the method, identifying an area within the first field of view of the obtained thermal image for subsequent monitoring comprises identifying a contiguous cluster of high-temperature pixels in the obtained thermal image that exhibits the highest temperature.

[0031] In one or more embodiments, the method, while in the third operating mode, includes shortening the second time intervals as the temperature of the identified area increases.

[0032] In one or more embodiments, the method, while in the fourth operating mode, comprises issuing a second alarm signal if a temperature in any area of ​​the range detected by the scans of the third type increases by a predetermined third value or more within a predetermined third number of successive scans of the third type or within a predetermined third time interval after the first alarm signal has been issued.

[0033] In one or more embodiments, the method comprises opening an aperture covering at least the thermal sensor arrangement and an associated optical arrangement upon entering the second operating mode and closing the aperture upon returning to the first operating mode.

[0034] In one or more embodiments, the method comprises performing a cleaning process for the thermal sensor arrangement and / or the associated optical arrangement at predetermined third time intervals and / or when differences between a thermal image acquired for an unoccupied parking space and a reference image exceed respective predetermined threshold values.

[0035] According to a second aspect of the invention, a system for detecting battery fires in electric or hybrid vehicles (EVs) in parking lots is presented. The system comprises a power supply, a vehicle presence sensor, a thermal sensor array, and an associated lens system, and may further include a temperature sensor and / or a humidity sensor. The system also includes a control unit with one or more microprocessors or corresponding logic circuits and associated volatile and non-volatile memory, as well as a communication interface, which may be wired or wireless. The non-volatile memory stores computer program instructions which, when executed by the one or more microprocessors of the control unit, configure the system to perform embodiments of the method according to the first aspect of the invention.

[0036] The methods described above can be represented by computer program instructions. According to a further aspect of the invention, a computer program product therefore comprises computer program instructions which, when executed by a microprocessor or corresponding logic circuits of a control unit of a system according to the second aspect of the invention, cause the microprocessor to execute embodiments of the method according to the first aspect of the invention and to control the hardware components of the system accordingly.

[0037] Computer program instructions or code for performing operations for implementations can comprise any number of lines and be written in any combination of one or more programming languages, including an object-oriented programming language such as Python, Ruby, Java, Smalltalk, C++, or the like, and traditional procedural programming languages ​​such as the programming language "C" or the like, and / or machine languages ​​such as assembly languages. The code can be executed entirely on the user's computer, partially on the user's computer as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on the remote computer or server.In the latter case, the remote computer can be connected to the user's computer via any type of network, including a local area network (LAN), a wireless LAN (WLAN) or a wide area network (WAN), or the connection can be made, for example, via the Internet using an Internet service provider (ISP) to an external computer.

[0038] Computer program instructions can be stored or transmitted on a computer-readable medium or storage medium. The medium or storage medium can be materially or physically embodied, for example, in the form of a hard drive, a solid-state drive, a flash memory device, an EPROM or EEPROM, or the like. However, the medium or storage medium can also comprise a modulated electromagnetic, electrical, or optical signal that is received by the computer via a suitable communication interface and transmitted to and stored in the computer's memory.

[0039] The described features, structures, or properties of the embodiments can be combined in any suitable way. This description provides numerous specific details, such as examples of programming, software modules, user selections, network transactions, database queries, database structures, hardware modules, hardware circuits, hardware chips, etc., to convey a comprehensive understanding of the embodiments. However, a person skilled in the art will recognize that embodiments can be realized without one or more of the specific details or with different methods, components, materials, etc. In other cases, known structures, materials, or processes are not shown or described in detail so as not to obscure aspects of an embodiment.References in this description to "an embodiment" or similar phrases mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment. Therefore, the phrase "in an embodiment" and similar phrases in this description may, but need not, refer to the same embodiment and may mean "one or more, but not all, embodiments," unless expressly stated otherwise. The terms "including," "comprising," "with," and variations thereof mean "including, but not limited to," unless expressly stated otherwise. An enumeration of elements does not imply that some or all elements are mutually exclusive, unless expressly stated otherwise.The terms "a", "an", and "that" also refer to "one or more" unless explicitly stated otherwise. A numbered list of items does not imply that some or all of the items are mutually exclusive unless explicitly stated otherwise. The terms "a" and "the" also refer to "one or more" unless explicitly stated otherwise.

[0040] When aspects of the embodiments in this description are described with reference to schematic flowcharts and / or schematic block diagrams of methods, devices, systems, and program products according to the embodiments, it is understood that each block of the schematic flowcharts and / or schematic block diagrams and combinations of blocks in the schematic flowcharts and / or schematic block diagrams can be implemented by code. This code can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to create a device such that the instructions executed by the processor of the computer or other programmable data processing device provide means for implementing the functions / actions specified in the flowcharts and / or block diagrams.

[0041] It should be noted that in some implementations or embodiments, the functions specified in the exemplary embodiments shown in the figures may occur in a different order than shown in the figures. For example, two blocks shown consecutively may actually be executed essentially simultaneously, or the blocks may sometimes be executed in reverse order, depending on the related functionality. Other steps and procedures may be conceived which, in their function, logic, or effect, correspond to one or more blocks or parts thereof shown in the figures.

[0042] The present invention reduces the risk of fire and the significant damage caused by a burning electric vehicle in a parking lot, including business losses due to building closures, damage to building foundations, and, in the worst case, even fatalities. In particular, the method and system according to the invention can detect abnormal temperature increases faster and earlier than conventional solutions. The sensor and method are capable of quickly detecting and sensing abnormal temperature increases in all types of electric vehicles, including hybrid electric vehicles.

[0043] It is obvious that the system can also be used in battery-powered trains, boats or airplanes.

[0044] Field tests have shown that battery overheating leading to a fire can be detected at least 3 minutes before the actual battery fire and about 10 minutes earlier than by a smoke detector. BRIEF DESCRIPTION OF THE DRAWING

[0045] In the following section, embodiments of the invention are described with reference to the accompanying drawings. The drawing shows Fig. an exemplary system according to the invention, which is installed at a parking space for a car, Fig. a schematic block diagram of an exemplary sensor system, Fig. an exemplary flowchart of a typical recognition process according to the present invention, Fig. an optional sub-process for lens cleaning, which is controlled by a watchdog timer, Fig. a sectional view of a parked car above a thermal overheating sensor according to the invention, Fig. a typical battery temperature curve when a malfunction occurs during the charging process, and Fig. an exemplary block diagram of a device set up for implementing embodiments of the method according to the invention.

[0046] The illustrations may use identical reference symbols for identical or similar elements or features. DETAILED DESCRIPTION OF THE EXECUTION FORMS

[0047] Fig. Figure 1 shows an exemplary system according to the invention, installed at a parking space for a car. All other parking spaces can be similarly equipped. A sensor system 200 is integrated into the ground 900. The sensor system 200 is connected to a central unit 600. Likewise, further sensor systems 201, 202 are connected to this central unit 600, one for each parking space. The central unit 600 can be connected to the sensor systems 200, 201, 202 and configured to transmit an alarm via the communication device 700. Alternatively, the central unit 600 can perform part of the processing of the sensor data, i.e., the thermal images, to evaluate an event of thermal runaway of the battery 810 of an electric vehicle 800 parked in the parking space. The central unit 600 can also be configured to manage the detection of a car parked in the parking space, e.g.,when the sensor systems 200, 201, and 202 are used both as a thermal sensor array and as a vehicle presence sensor. The central unit 600 or the sensor system 200 can also be configured for communication with a charging device assigned to the parking space (not shown in the figure) or with a safety switch coupled to selectively interrupt the power supply to the charging device (not shown in the figure). The connection 210 between the sensor systems can be wired or wireless. Wireless connections can include any low-power transmission system, such as Bluetooth LE, LoRaWAN, etc. In the case of wireless transmission, the sensor system 200 can be self-powered by a battery system and / or an energy harvesting system, or it can be powered via cable.

[0048] Fig. Figure 1 shows a schematic block diagram of an exemplary sensor system 200. The local battery 250 supplies the entire sensor system 200 with energy. The battery 250 can be of the primary or secondary type. In the case of a secondary battery, an energy harvesting circuit can be provided for charging the battery 250. Various types of energy harvesting circuits are generally known and can include, among others, a photovoltaic cell or an electromagnetic system that absorbs energy from the electromagnetic field of nearby charging currents. Thermal generators that utilize a temperature difference to generate electrical energy are also conceivable. An energy management system 220, controlled by a control unit 300, regulates the energy supply of the various system components. Depending on the type of control, the energy management system 220 can be integrated into the control unit 300.A vehicle presence sensor 400 is provided, which can be optical, thermal, radar, lidar, or of any other suitable type. The vehicle presence sensor 400 can generally be of a conventional design and type, such as those already used in existing parking systems. Other alternatives, such as an induction loop or capacitive systems, are not excluded. As shown later in the flowchart, the vehicle presence sensor 400 detects the presence of a vehicle and triggers the operation of the sensor system 200.

[0049] The sensor system 200 further includes a lens system 420, which is required to control the viewing angle of the thermal sensor arrangement 440, i.e., to widen and / or narrow the viewing angle. Due to the small viewing distance between the underside of a vehicle and the surface of the parking space, which is typically in the range of about 10 to 20 cm, a viewing angle of almost 180 degrees in all directions must be provided, i.e., nearly a hemisphere or a semi-ovoid. This can be achieved by a conventional lens system or a Fresnel lens, by a prism system, or by a scanning mirror system.

[0050] A thermal sensor assembly 440 is provided, which is essential for measuring the temperature of the underside of a vehicle that is closest to the space in which the vehicle battery 810 is mounted. It can be advantageous if the sensor has more than one pixel and if these pixels can be selectively activated via the control unit 300. It is evident that instead of one large thermal sensor assembly, several subunits can be used, which can be distributed throughout the parking area. Fig. Figures 430, 432, and 434 show an example of different viewing angles of a single thermal sensor arrangement, which can be selected by adjusting the lens system, by selecting a subset of pixels from the totality of pixels, by adjusting a movable scanning system, or the like. Fig. This also shows that monitoring the battery temperature may only require a portion of the pixels of a multipixel thermal sensor assembly 440. The unused pixels or parts of the multipixel thermal sensor assembly can be switched off to save energy. A temperature sensor 450 and a humidity sensor 480 can be provided to supply a reference value for the air temperature and humidity in the vehicle's environment. Air temperature and humidity are important parameters for fully capturing the accurate temperature gradient of a cooling or heating battery. This measurement does not need to be precise, and the sensors could be an integrated sensor on a printed circuit board containing the system's various electronic components.

[0051] A shutter 490 allows the optical systems of the sensor system 200 to be covered when not in use or when the parking area surface is being cleaned. The shutter 490 can also include a cleaning mechanism for cleaning the lens system 420. For this purpose, a movable part of the shutter 490, such as a sliding cover or similar, can be equipped with a wiper, brush, or similar device that cleans the lenses when the shutter 490 opens or closes.

[0052] Fig. Figure 1 shows an exemplary flowchart of a typical recognition process according to the present invention. Block 901 represents an energy-saving mode for conserving energy provided by a battery and / or by energy harvesting. Two functions can be performed in energy-saving mode: - a lens cleaning cycle 905, which is started by the watchdog timer 903. The lens cleaning cycle is described further below with reference to Fig. described. - a vehicle detection and monitoring loop comprising a detection step 921 and a decision step 923.

[0053] In the vehicle detection and monitoring loop, detection step 921 serves to identify any vehicle parked in a previously vacant parking space. This step can involve the use of a vehicle presence sensor 400, which can be of a conventional design and functions similarly to sensors already known from parking management systems. The vehicle presence sensor 400 detects whether a vehicle is above or near the sensor. Using a separate vehicle presence sensor can have the advantage of consuming less power to detect a vehicle than the thermal sensor assembly 440.

[0054] If a vehicle is detected (yes branch of decision step 923), the system switches to a second operating mode. In step 925, the shutter covering the thermal sensor array 440 is opened, and then in step 927, one or more type 1 scans are performed by the thermal sensor array 440. These type 1 scans 927 can be performed continuously or at intervals, e.g., every few minutes. To limit energy consumption, the entire field of view 430 of the thermal sensor array 440 is evaluated. The type 1 scan is also referred to here as a micro-power scan. An area, e.g., a group of contiguous pixels, with the highest temperature is identified and selected. Decision step 929 determines whether the parked vehicle is equipped with an internal combustion engine or an electric drive.

[0055] Detecting the warmest region, the hotspot, i.e., a group of pixels with the highest brightness in a thermal image captured by the thermal sensor array, may involve performing a threshold decision to divide the pixels into bright and dark pixels, and subsequently performing connected-component analysis (CCA) to identify and extract connected bright pixel areas.Other methods that can be used in conjunction with the present invention to identify the area of ​​highest temperature in the thermal image include blob detection using convolution, superpixel detection by applying algorithms such as quickshift image segmentation or watershed segmentation, application of a DBSCAN clustering algorithm that groups together data points that are close to each other and have a high density of neighboring points, while points in low-density regions are treated as noise, and / or application of connected component labeling (CCL), a fundamental algorithm in image processing that assigns a unique number to each connected component of a binary image. Other methods that can be used here include the use of artificial intelligence (AI) to detect and select the area of ​​highest temperature in the thermal image, e.g.,Using AI image classification and implementing a Convolutional Neural Network (CNN), the largest contiguous high-temperature cluster is selected based on the number of connected pixels and used for further monitoring.

[0056] As mentioned above, to increase the reliability of the decision, two or more consecutive thermal images can be acquired and analyzed at suitable time intervals, e.g., one minute. The interval between these times can depend on the ambient temperature and increase with rising ambient temperature. The two or more consecutive images are then compared, and areas where the temperature has changed between two thermal images are identified.

[0057] Internal combustion engines and their exhaust gases are easily identifiable based on their temperature profile and shape in the thermal image. After the vehicle is parked, temperatures exceeding 100 °C may occur, but these can drop rapidly by several degrees Celsius within ten minutes. If such a profile is detected, the system can switch to a mode that simply detects whether this particular vehicle is still present, for example, for an optional parking management system (not shown in the figure). This can be done using the vehicle presence sensor 400. Thus, if a vehicle with an internal combustion engine is detected (decision step 929, branch "No"), the gate in step 926 can be closed, and the procedure returns to block 901 for the energy-saving mode.

[0058] If an EV is detected, the "Yes" branch is selected in step 929, and a third operating mode is invoked. This third operating mode involves performing a regular scan, also known as a low-power scan, in step 931. The low-power scan can be performed at initial time intervals, for example, every five to ten minutes. The low-power scan only monitors the area identified as being connected to a battery, as shown in Fig. , Viewpoint 432, is shown. Next, based on the results of successive low-power scans, Decision Step 933 determines whether the battery temperature is changing in a way that indicates an abnormal situation. If the difference between the vehicle body temperature and the ambient temperature does not exceed a predetermined initial value, e.g., 10 °C, and / or if the battery temperature continues to decrease, the procedure proceeds to Step 935 after the "No" branch of Step 933 has been completed. It should be noted that the battery temperature can be approximately 20 °C to 30 °C above the ambient temperature that can be detected by Temperature Sensor 450. Step 935 simply determines whether the vehicle is still parked. If so, the "Yes" branch of Step 935, the procedure continues to perform low-power scans at intervals.In the negative case, the "no" branch of step 935, the procedure returns to block 901 for the low-power mode and closes the shutter in step 926. It should be noted that further checks for temperature hotspots can be performed during the third operating mode.

[0059] If, in decision step 933, it is determined that the difference between the vehicle's body temperature and the ambient air temperature exceeds the specified first value or the battery temperature increases (yes branch of step 933), the procedure switches to a fourth operating mode and proceeds to step 940.

[0060] In step 940, third-type scans are performed at second time intervals, which are shorter than the first time intervals, e.g., less than one minute. Entering this operating mode, also referred to here as fourth mode, signifies a high risk of an incident, and the system therefore monitors the battery more frequently and with higher spatial resolution. The higher spatial resolution enables the detection and tracking of smaller hotspots, thus providing more detailed information. At this point, artificial intelligence (AI) can assist the monitoring through the use of 2D deep learning algorithms. Depending on the implementation, these rather energy-intensive algorithms can be executed outside the vehicle detection system, and the third-type scans can be transmitted to an external server in step 942, which returns the results to decision step 944.It should be noted that the second time intervals can be automatically adjusted, i.e., shortened, if necessary, and that the spatial resolution can also be adjusted if necessary. Decision step 944 checks whether the temperature in any area of ​​the area covered by the third-type scans drops within the predetermined first number of consecutive third-type scans or within the predetermined first time period. If the result is positive, the "yes" branch of step 944, the procedure continues with step 931, effectively returning to the third-type operating mode. If an alarm was previously triggered in step 948, the alarm can be cleared in step 948b.In the negative case, the "No" branch of step 944, the procedure checks in step 946 whether the temperature in any part of the area covered by the third-type scans increases by more than a predetermined second value within a predetermined second number of consecutive third-type scans or within a predetermined second time period. An example critical temperature change is 1 °C / minute over a period of 5 minutes. In the negative case, "No" branch of step 946, the procedure returns to step 940 and continues the third-type scans. In the positive case, "Yes" branch of step 946, a first alarm signal is issued in step 948, and the procedure returns to step 940 and continues the third-type scans.It should be noted that the alarm can be repeated or sustained if the loop consisting of steps 940, 942, 944, 946, and 948 detects that the temperature continues to rise. Alternatively, the alarm can be triggered only on the first attempt. The alarm and its clearing can be transmitted via wired or wireless connection to a monitoring system and / or a human monitor.

[0061] Fig. Figure 905 shows an optional lens cleaning subprocess, controlled by the watchdog timer 903. After a predetermined time interval, e.g., once a day, the watchdog timer 903 triggers the lens cleaning loop 905. In step 907, after the shutter 490 opens, a thermal image is acquired and compared to a reference image retrieved from memory. The reference image, which can be acquired during system installation or at a later time, e.g., after maintenance or at regular intervals, can be selected according to the current ambient temperature and / or humidity. If the images are sufficiently similar (yes branch of step 907), the subprocess ends and returns to the watchdog timer loop 903.If the images differ too much, the "No" branch of step 907 assumes that the lens system 420 is dirty, and a lens cleaning process is performed in step 911. The lens cleaning may involve repeatedly opening and closing the shutter 490. After the lens cleaning process is complete, step 913 verifies whether the lens cleaning was successful, for example, by repeating the comparison with a current thermal image and a reference image. If successful (Yes branch of step 913), the procedure returns to the watchdog timer loop 903. If not successful (No branch of step 913), a maintenance alarm is issued in step 915, and the procedure returns to the watchdog timer loop 903. Alternatively, the lens cleaning process can be repeated in step 911.

[0062] Fig. Figure 440 shows a cross-section of a parked car above a sensor for detecting a thermal runaway battery according to the invention, and in particular the optical situation of the thermal sensor arrangement 440 together with the lens system 420. All viewing angles can be 1- or 2-dimensional, depending on the pixel array of the thermal sensor. If the sensor is only 1-dimensional, the lens system must adjust the viewing angle so that most of the car body is captured. 1-dimensional viewing angles may require the sensor to scan an area. The viewing angle 430, the "full viewing angle" of the sensor, covers most of the underside of the vehicle. Due to the extreme ratio between the vehicle's ground clearance and the length of the car body, the full viewing angle can be up to 175°.It is important that the entire underside of the vehicle is scanned, as the battery may be located in different places or installed in two or more locations across multiple sections. Viewpoint 432 covers the area identified as the battery compartment. This area is detected by varying temperatures and / or temperature changes, i.e., temperature profiles over time. Viewpoint 434 represents a viewpoint for an identified hotspot. This hotspot is only detected when a thermal runaway begins and serves for more detailed monitoring.

[0063] Fig. This shows a typical battery temperature curve when a malfunction occurs during charging. The x-axis is linear and represents time, while the y-axis is logarithmic and represents temperature. It should be noted that similar malfunctions can also occur during battery use or while the battery is parked after use. However, these latter malfunctions are difficult to provoke for testing purposes, making measurements for illustrative purposes hard to obtain.

[0064] The charging process begins at T1, assuming the battery has an ambient temperature of approximately 16 °C. The temperature rises slowly due to chemical processes occurring within the battery and some resistance losses. At T2, the battery temperature has already risen to 18 °C. At T3, the temperature increase reaches a threshold of "1 °C / minute for more than 5 minutes," indicated by the reference number 120. At T4, the battery experiences thermal runaway; the temperature rises rapidly to over 1000 °C. At T5, the battery is effectively engulfed in flames, burning parts of the car or oily or other flammable liquids, and emitting smoke that can be detected by a conventional smoke detector. The heat is not yet sufficient to trigger a conventional sprinkler system heat capsule positioned above the vehicle, as the flames have not yet reached that far.

[0065] The method according to the invention, after detecting that a car has been parked, performs a first-type scan at T1 in the second operating mode. The method then switches to the third operating mode at T2 and to the fourth operating mode at T3. Experiments have confirmed that the system according to the invention triggers an alarm approximately 10 minutes earlier than a conventional smoke detector, which only responds at T5.

[0066] Fig.Figure 1 shows a schematic block diagram of an exemplary control unit 300 of the sensor system 200 according to the invention. The control unit 300 comprises interfaces 302 for controlling various system components such as the shutter 490, the thermal sensor arrangement 440, the lens system 420, and the like, as well as for transmitting alarms to other systems. The control unit further comprises one or more microprocessors 306 and associated volatile 308 and non-volatile memories 310. The various components and elements of the control unit 300 are interconnected via one or more data and / or signal lines or buses 312. The non-volatile memory 310 stores computer program instructions which, when executed by the one or more microprocessors 306, configure the control unit 300 to implement or execute embodiments of the method according to the first aspect of the invention as described herein. LIST OF REFERENCE MARKS (PART OF THE DESCRIPTION) 100 procedures 110 temperature 112 Ambient temperature Exceeded 120°C / min for >5 min. 200 sensor system 201 Sensor system 202 Sensor system 210 Power supply 220 Power Management 240 Energy recovery circuit 250 battery 260 wireless communication 280 antenna 300 control unit 302 interface(s) 304 interface(s) 306 microprocessor(s) 308 volatile memory 310 non-volatile memory 312 Data / signal line / bus 340 drivers 360 LED 400 vehicle presence sensors 420 lens system 430° viewing angle of vehicle underbody 432 viewing angle battery 434 viewpoints “Hotspot” 440 thermal sensor array 450 temperature sensor 480 humidity sensor 490 Closure 500 cases 600 Central unit 700 communication equipment 800 vehicles 810 battery 850 parking spaces 900 floor 901 low-power mode 903 Watchdog timer loop 905 905 Lens cleaning loop 907 Comparison of current thermal image with reference 909 911 Lens cleaning process 913 Lens cleaning successful? 915 Maintenance Alarm 921 Vehicle Identification 923 Decision step 925 Open the closure 926 Close the clasp 927 Micro-Power Scan via thermal sensor array, limited pixel count 929 Choice of combustion engine or electric drive 931 Low-Power Scan 933 Decision between normal or excessive heating 935 Vehicle still present? 940 Scan of the third kind 942 Signal transmission to external server 944 Alarm threshold (1 °C / minute for more than 5 minutes) 946 Check temperature gradients 948 Trigger alarm 948b Clear alarm 880 Parking Management System

Claims

[1] Method (100) for detecting battery fires of an electric or hybrid vehicle, EV, (800) in a parking lot (850), comprising in a first operating mode: - Detect (921, 923) whether the occupancy status of the parking space (850) changes from empty to occupied, and generate a corresponding signal, and furthermore comprehensively, in a second operating mode, which is called after receiving the signal indicating that the occupancy status of the parking space (850) has changed from empty to occupied: - Performing (927) one or more scans of a first type via a thermal sensor arrangement (440) over a first field of view (430) to obtain a corresponding thermal image of a first type, and evaluating the obtained thermal image of the first type to identify related pixel clusters with high temperature therein, - Determine (929) whether a parked vehicle is an EV (800), and in the negative case: - Return to the first operating mode until the occupancy status of the parking space (850) changes again from empty to occupied, and if so, call up a third operating mode which includes the following: - Identifying an area (434) within the first field of view (430) of the received thermal image for subsequent monitoring, - Performing (931) second-type scans at first time intervals over a second field of view (432) which is smaller than the first field of view (430) and includes the identified area (434), as long as the temperature of the identified area (434) does not increase, as long as the temperature in the second field of view (432) outside the identified area (434) is not higher than the ambient temperature by a predetermined first value, and as long as the parking space (850) remains occupied, wherein the method (100) comprises calling up a fourth operating mode if the temperature of the identified area (434) increases or if the temperature in the second field of view (432) outside the identified area (434) exceeds the ambient temperature by more than the predetermined first value while the parking space (850) remains occupied, wherein the fourth operating mode comprises: - Performing (940) scans of a third type, covering at least the identified area (434), at second time intervals shorter than the first time intervals, while a temperature in any part of the area covered by the scans of the third type is rising or does not fall within a predetermined first number of successive scans of the third type or within a predetermined first time span, and while the parking space (850) remains occupied, and - Issuing (946) a first alarm signal when a temperature in any area of ​​the area covered by the third-type scans increases by more than a predetermined second value within a predetermined second number of successive third-type scans or within a predetermined second period, or Return to the third operating mode if the temperature in any area of ​​the area covered by the third-type scans falls within the predetermined first number of consecutive third-type scans or within the predetermined first period while the parking space (850) remains occupied. [2] Method (100) according to claim 1, wherein the change in the occupancy status of the parking space (850) from empty to occupied and / or the continued occupancy of the parking space (850) is indicated by a signal received by an independent parking monitoring system or a corresponding sensor of a system (200) implementing the method (100), and / or by a signal received by a process that evaluates successive scans of the first, second, third and fourth type in the respective operating mode. [3] Method (100) according to claim 1 or 2, wherein determining (929) whether the parked vehicle is an electric vehicle (800) comprises identifying and evaluating a position of one or more connected high-temperature pixel clusters within the first field of view (430) of the obtained thermal image by evaluating temperature changes of the connected pixel cluster over a number of successive scans of the first type or a predetermined third period and / or by receiving a signal indicating that a charging device connected to the parking space (850) is activated. [4] Method (100) according to one or more of the preceding claims, wherein identifying the area (434) within the first field of view (430) of the obtained thermal image for subsequent monitoring comprises identifying a contiguous high-temperature pixel cluster in the obtained thermal image with the highest temperature. [5] Method (100) according to one or more of the preceding claims, further comprising that during the third operating mode the second time intervals are shortened as the temperature of the identified area (434) increases. [6] Method (100) according to one or more of the preceding claims, further comprising that in the fourth operating mode a second alarm signal is issued when a temperature in any area of ​​the area detected by the scans of the third type increases by a predetermined third value or more within a predetermined third number of successive scans of the third type or within a predetermined third time period after the first alarm signal has been issued. [7] Method (100) according to one or more of the preceding claims, further comprising that upon entering the second operating mode an aperture (925) is opened which covers at least the thermal sensor arrangement (440) and an associated optical arrangement, and that the aperture is closed upon returning to the first operating mode. [8] Method (100) according to one or more of the preceding claims, further comprising performing a cleaning process for the thermal sensor arrangement (440) and / or the associated optical arrangement at predetermined third time intervals and / or when differences between a thermal image taken for an unoccupied parking space (850) and a reference image exceed respective predetermined threshold values. [9] A system (200) for detecting battery fires of electric or hybrid vehicles (EV) (800) in parking lots (850), wherein the system (200) comprises a power supply (210), a vehicle presence sensor (400), a thermal sensor arrangement (440) and an associated lens system (420), a temperature sensor (450), a humidity sensor (480) and a control unit (300) with one or more microprocessors (306) and associated volatile (308) and non-volatile memory (310), wherein the non-volatile memory (310) stores computer program instructions which, when executed by the one or more microprocessors (306) of the control unit (300), configure the system (200) to execute the method (100) according to one or more of claims 1 to 8. [10] System (200) according to claim 9, wherein the power supply (210) comprises a battery (250) and wherein an energy recovery circuit (240) is provided for charging the battery (250) or for maintaining a battery charge. [11] Computer program product comprising computer program instructions which, when executed by a microprocessor (306) of a system (200) from one or more of claims 9 or 10, cause the microprocessor (306) to execute embodiments of the method (100) according to one or more of claims 1 to 8 and to control the hardware components of the system (200) accordingly. [12] Computer-readable medium or data carrier that reproducibly transmits or stores a computer program product according to claim 11.

Images