System and method for early detection of thermal runaway from the battery

WO2025188255A8PCT designated stage Publication Date: 2025-10-02ELOC8 SRO
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/SK2025/000005
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2024-03-06
Filing Date
2025-03-05
Publication Date
2025-10-02

AI Technical Summary

Technical Problem

Existing technologies for detecting thermal runaway in lithium batteries, particularly in aviation, do not adequately account for volatile organic compounds and are not sufficiently suitable for aviation applications, posing a hazard due to high temperatures, pressure, and gas release.

Method used

A system utilizing gas concentration sensors, temperature sensors, a microcontroller unit with an advanced machine learning algorithm, wireless communication, and cloud storage, which monitors volatile organic compounds and temperature changes to detect thermal runaway using IoT technology.

Benefits of technology

Enables early and reliable detection of thermal runaway in lithium batteries by accurately monitoring gas concentrations and temperature fluctuations, providing timely warnings and reducing potential hazards.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure SK2025000005_02102025_PF_FP_ABST
    Figure SK2025000005_02102025_PF_FP_ABST
Patent Text Reader

Abstract

The early detection system of thermal runaway from a battery, containing an air cargo (1) with a detected battery (11), a gas concentration sensor (2) and / or a temperature sensor (3), microcontroller unit (4) with an advanced machine learning algorithm, wireless communication (5), an alarm (6) and a cloud storage (7) with a database, using Internet of Things technology, characterized in that the microcontroller unit (4) with an advanced machine learning algorithm, located within the range of the detected battery (11) air cargo (1) is connected to at least one gas concentration sensor (2) and / or at least one temperature sensor (3), wherein the microcontroller unit (4) with an advanced machine learning algorithm is connected by wireless communication (5) to the alarm (6) and a cloud storage (7) with a database.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] Name of the invention:

[0002] System and Method for Early Detection of Thermal Runaway from the Battery

[0003] Technical Field

[0004] The invention belongs to the field of measurement of physical quantities within physics, monitoring and control of batteries within electrical engineering, and testing of aircraft component inspection within aviation, and relates to a system and method for monitoring the state of lithium batteries with the ability to detect and provide advance warning of thermal runaway and thus of a possible impending hazard.

[0005] Background

[0006] Lithium batteries are part of the equipment used by passengers, crew, as well as transported luggage and other cargo in aviation. One of the main risks associated with lithium batteries is thermal runaway. Thermal runaway is a phenomenon in which a lithium cell enters an uncontrollable, self-heating state. It can result in extremely high temperatures and pressure, the rapid release of flammable gases from the cells, followed by smoke and fire. This creates a potential hazard to passengers, crew members, and the aircraft.

[0007] The aforementioned shortage of lithium batteries is also known in the operation of electric vehicles. In the state of the art, a solution according to EP 3 843 195 "THERMAL RUNAWAY DETECTING DEVICE, BATTERY SYSTEM, AND THERMAL RUNAWAY DETECTING METHOD OF BATTERY SYSTEM", the core of which is to compare the results of the measured voltage and temperature values of two measurement circuits, the first from the contacts of a single battery cell and the second from the contacts of the entire battery, and by comparing the results in the controller, evaluate whether thermal runaway is occurring. The above solution does not take into account the leakage of volatile organic compounds into the air.

[0008] A solution according to EP 3 800 725 entitled "THERMAL RUNAWAY DETECTION METHOD, DEVICE AND SYSTEM FOR BATTERIES, AND BATTERY MANAGEMENT UNIT" comprising: obtaining an output signal of an air pressure sensor included in the battery pack, and obtaining information about parameters of the battery pack; determining information about the status of the air pressure sensor based on the output signal of the air pressure sensor. After evaluating the information obtained, the system evaluates the possibility of thermal runaway.

[0009] The WO disclosure 2023006698 entitled "DETECTION SYSTEM AND METHOD" is used by a system comprising a coating applied to the outside of one or more battery cells, wherein said coating is selected to decompose at a temperature range useful for said detection and so as to release a detectable volatile compound. A solution according to US 20200348365 entitled "Thermal runaway detection circuit and method" includes: a sensing module comprising a sensing cable; a detection module coupled to the sensing cable and comprising at least one set of voltage dividing resistors; a processing module connected to the sensing module, wherein the processing module is configured to receive thermal runaway detection data and determine whether thermal runaway is occurring. Another known solution is from US 20210111443, titled “METHOD, APPARATUS, SYSTEM FOR DETECTING BATTERY THERMAL RUNAWAY, AND unitBATTERY MANAGEMENT UNIT”, The method comprises obtaining an output signal of an air pressure sensor located in a battery pack and obtaining information about parameters of the battery pack. The evaluation of this data leads to a conclusion as to whether or not thermal runaway of the battery pack is occurring. The above solutions which are part of the state of the art relate to thermal runaway of automotive batteries and are not sufficiently suitable for the determination of thermal runaway in aviation.

[0010] Summary of the Technical Solution

[0011] The disadvantages of the prior art are partially eliminated by a system and method for early detection of a thermal runaway battery according to the present invention comprising an air cargo with a detected battery, a gas concentration sensor and / or a temperature sensor, a microcontroller unit with an advanced machine learning algorithm, and wireless communication, an alarm and a cloud storage with a database, utilizing Internet of Things technology, wherein the principle of the loT technology is that a microcontroller unit with an advanced machine learning algorithm located within range of the detected air cargo battery is coupled to at least one gas concentration sensor and / or at least one temperature sensor. The microcontroller unit with the advanced machine learning algorithm is connected by wireless communication to the alarm and to the cloud storage with the database.

[0012] Solutions for the early detection of thermal runaway from a battery that use a gas substances concentration sensor with detection of the following configurations are advantageous:

[0013] • volatile organic compounds and hydrogen fluoride

[0014] • volatile organic compounds and carbon monoxide

[0015] • volatile organic compounds and hydrocarbons

[0016] • volatile organic compounds and ethylene

[0017] • volatile organic compounds and molecular hydrogen

[0018] • hydrogen fluoride, carbon monoxide and carbon dioxide

[0019] • molecular hydrogen, carbon monoxide and carbon dioxide

[0020] • hydrocarbons, carbon monoxide and carbon dioxide

[0021] • ethylene, carbon monoxide and carbon dioxide

[0022] • volatile organic compounds, carbon monoxide and carbon dioxide.

[0023] Volatile organic compounds in the context of this technical solution are mainly carbon dioxide, mold, volatile compounds such as gasoline, varnishes and oils in spray.

[0024] Also advantageous are solutions for the early detection system of thermal runaway from the battery, which utilize a high-precision temperature sensor with an accuracy of + 1.5% of the measured values, or an advanced temperature sensor with an accuracy of + 0.5%.

[0025] The shortcomings mentioned in the prior art are also partially solved by the method of early detection of thermal runaway from the battery, the principle of which lies in the fact that information flows with data on gas concentration values and / or with data on temperature values, measured using gas concentration sensors and / or using temperature sensors, evaluated by a microcontroller unit with an advanced machine learning algorithm, are sent by wireless communication to the alarm and to a cloud storage with a database. The cause of the thermal runaway hazard is a failure of a lithium-ion battery. This failure can be caused by an internal failure or external conditions. One example of such an internal failure is an internal short circuit In a lithium cell, the cathode and anode electrodes are physically separated by a component called a separator. Failures in the cell that compromise the integrity of the separator can cause an internal short circuit, which can result in thermal runaway. This is particularly likely in low-quality battery cells.

[0026] Examples of failures caused by external conditions include: overcharging (which can be caused by incompatibility between the cell and charger or a poorly designed battery management system (BMS); multiple overdischarges and subsequent recharges; repeatedly discharging a cell or battery below the lower voltage limit recommended by the cell manufacturer and then recharging the cell.

[0027] The issue of thermal runaway is addressed by several recognized institutions, UL (Underwriters Laboratories, Inc., 1894) - an independent certification organization, FAA (Federal Aviation Administration, 1958) - an agency of the US Department of Transportation, EASA (European Aviation Safety Agency, 2002) - the European Union Agency for Aviation Safety, as well as many world universities.

[0028] Tests with the induction of various above-mentioned battery failures, including the stage of battery destruction, have proven that early detection of thermal runaway from a lithium battery can be achieved by monitoring the immediate environment of the battery, by monitoring temperatures and evaluating the air surrounding the battery for the presence of the concentration of hydrogen (H2), carbon dioxide (CO2), carbon monoxide (CO) as volatile organic compounds - VOC.

[0029] By monitoring the concentration of total volatile organic compounds (TVOC), it is possible to assess the estimated concentration of eCO2, the estimated (equivalent) CO2.

[0030] More information on the issue of thermal runaway from batteries can be found at the following resources: https: / / www.fire.tc.faa.gov / pdf / tc20-12.pdf https: / / www.fire.tc.faa.gov / CargoSafety / Hazards / 1001-Passenger Baggage https: / / www.fire.tc.faa.gov / 2022Conference / files / Cabin Flight Deck Fire Protection IH / PashaTripDatabasePED / Pasha TRIPDatabasePED Pres.pdf https: / / www.fire.tc.faa.gov / pdf / TC-15-59.pdf https: / / ul.org / research / electrochemical-safety / getting-started-electrochemical-safety / what -causes45thermal#:~:text=Thermal%20runaway%20is%20a%20phenomenon,cell%20venting% 2C%20smoke%20and%20fire. https: / / flightsafety.org / asw / mar08 / asw mar08 p42-47.pdf

[0031] According to FAA tests, the most significant gases causing thermal runaway are total hydrocarbon emissions (THC), hydrogen, carbon monoxide, and carbon dioxide. Other sources also report the presence of hydrogen fluoride and sound frequencies.

[0032] Early detection of thermal runaway from a battery is reliable using hydrogen, VOC, and temperature measurements along with an advanced machine learning algorithm focused on the angle of elevation of the measured values. Overview of images in drawings

[0033] The system and method for early detection of thermal runaway from a battery is explained in more detail with examples of implementation and with the help of drawings, which show:

[0034] Fig. 1 - schematic representation of the system for early detection of thermal runaway from a battery located in a ULD cargo container,

[0035] Fig. 2 - schematic representation of the system according to Fig. 1 with communication using wireless LoRa technology, the LoRaWAN protocol (loT network and protocol for the Internet of Things) and to cloud storage with a range of up to 5 km,

[0036] Fig. 3 - schematic representation of the system according to Fig. 1 with communication using wireless BLE technology (Bluetooth low energy) to a smartphone and to cloud storage with a range of up to 400 meters,

[0037] Fig. 4 - schematic representation of the system according to Fig. 1 with communication using wireless LTE Cat 1 bis technology (mobile network for the Internet of Things, 4G) and to cloud storage with high coverage by the LTE mobile network,

[0038] Fig. 5 - flow chart showing the individual steps of the method for early detection of thermal runaway from a lithium battery.

[0039] Examples of an implementation of an invention

[0040] Example 1

[0041] BASIC WARNING

[0042] The example presented presents a solution for early detection of battery thermal runaway using air temperature within air cargo (1) - pallets and containers known as ULD (Uniform Load Container), FRC (Fire Resistant Cargo Container) or other forms of storage, handling or transport of air cargo or cabin.

[0043] The basic warning system uses standard temperature sensors (3) and is aimed at monitoring sudden temperature increases.

[0044] The firmware in the main microcontroller unit (4) of the system is an advanced machine learning algorithm that continuously records data points that determine the normal operating temperature of the cargo. In the event of a thermal runaway, the temperature in the detected area will increase, which the system detects and alarms via BLE (52), LoRaWAN (51) or LTE Cat1 bis (53) communication technology. BLE communicates to mobile devices, LoRaWAN communicates to an anchor (gateway) located at the airport or on the aircraft, LTE Cat 1 bis communicates with the transmitters of the mobile operator providing LTE (4GF) services.

[0045] To ensure accurate detection, high-precision temperature sensors (3) are used, which allow for better differential monitoring between the calculated normal standard set by the firmware and the actual state. If the temperature in the ULD exceeds 50 °C, the system provides an early warning of smoke or fire.

[0046] Example 2

[0047] STANDARD WARNING

[0048] The example presents a more advanced solution for early detection of battery thermal runaway using air temperature within air cargo (1) ULD, FRC or other forms of air cargo storage, handling or transport or cabin space.

[0049] The standard warning utilizes advanced temperature sensors (3) and VOC (Volatile Organic Compounds) concentration sensors (2), which are released into the air as pollutants by evaporation, along with a combination of advanced hydrogen gas sensors, and is designed to monitor sudden changes in air concentration and temperature increases.

[0050] The firmware in the main microcontroller unit (4) of the device contains an advanced machine learning algorithm that continuously records data points that determine normal operating air concentration and cargo temperature. In the event of a thermal runaway, the air concentration and temperature suddenly change, which the system detects and alarms via BLE (52), LoRaWAN (51) or LTE Cat 1 bis (53) communication technology. BLE communicates to mobile devices, LoRaWAN communicates to an anchor located at the airport or in the aircraft, LTE Cat 1 bis communicates with the transmitters of the mobile operator providing LTE (4G) services.

[0051] To ensure accurate detection, high-precision sensors are used, including infrared - NDIR (Nondispersive Infra-Red) or laser sensors, which can recognize the difference in monitoring between the calculated normal standard set in the firmware and the actual state.

[0052] Depending on the application, the following combinations of temperature sensors (3) and gas concentration sensors (2) are used: temperature, volatile organic compounds and hydrogen fluoride temperature, volatile organic compounds and carbon monoxide temperature, volatile organic compounds and hydrocarbons temperature, volatile organic compounds and ethylene.

[0053] Example 3

[0054] ADVANCED WARNING

[0055] The example addresses early detection of battery thermal runaway using air concentration and temperature within an air cargo (1 ) ULD, FRC or other form of air cargo storage, handling or transport or aircraft cabin.

[0056] The advanced warning utilizes advanced temperature sensors (3) along with a combination of advanced gas concentration sensors (2) and thermal imaging cameras and is designed to monitor sudden changes in air concentration, thermal radiation and temperature increases.

[0057] The firmware in the main microcontroller unit (4) of the system is an advanced machine learning algorithm that constantly records data points that establish normal operating air concentration in the cargo, thermal radiation and temperature. In the event of a thermal runaway, the air concentration, thermal radiation and temperature in a cargo, warehouse or other container suddenly change, which the system detects and alarms via BLE (52), LoRaWAN (51) or LTE Cat 1 bis (53) (4G network) communication technology. BLE communicates to mobile devices, LoRaWAN communicates to an anchor located at the airport or on the aircraft, LTE Cat 1 bis communicates with the transmitters of the mobile operator providing LTE (4G) services.

[0058] To ensure accurate detection, high-precision sensors are used, including a thermal camera, non-scattering infrared sensors and laser particle detection, which can recognize the difference between the calculated normal standard set by the firmware and the actual state. Depending on the application, the following combinations of temperature detection and gas detection are used: temperature and hydrogen fluoride, carbon monoxide, carbon dioxide temperature and molecular hydrogen, carbon monoxide, carbon dioxide temperature and hydrocarbons, carbon monoxide, carbon dioxide temperature and ethylene, carbon monoxide, carbon dioxide temperature and volatile organic compounds, carbon monoxide, carbon dioxide.

[0059] Example 4

[0060] INTELLIGENT WARNING

[0061] The example presents a system and method for early detection of thermal runaway using advanced data analysis software.

[0062] To analyze the necessary parameters, data is obtained from all devices adopted within the invention and the examples, local weather and geographical information, and the system performs continuous analysis of all data. This analyzed data creates unique data points and limits, which are then communicated to individual devices over the air (and via a local network) to assist the device with accurate data points and limits to compare with the actual device data. Using these intelligent data points, the system algorithm can incorporate this additional data into its continuous analysis cycle, allowing the system to detect even greater differential changes in the environment. By incorporating intelligent warning data into any device, the accuracy of the system is increased.

[0063] FLOW CHART

[0064] The method of early detection of thermal runaway from the battery is shown in the flow chart in Fig. 5. The individual steps are as follows.

[0065] Start block - this is where the logic starts after the system starts.

[0066] Scanning (measuring the VOC quantity) is performed every 5 minutes. It is determined whether the measured VOC value exceeded 20 PPB (Parts per billion). If so, the logic continues and the system measures, in addition to VOC, H2 (hydrogen concentration) and temperature. These values are saved once as a baseline used for comparison in further decision logics. After 2 minutes, these parameters are measured again, the decision logic is performed to see whether the current vaiue for all 3 parameters is at least 1% greater than the baseline values. If not, the program logic returns to the beginning, scanning VOC in 5-minute intervals.

[0067] If the increase is above 1%, the first type of possible danger warning is sent (not a critical warning, but a possible anomaly, or even a thermal runaway from the battery at a very early stage), but it does not require immediate attention from a danger perspective, although the employee can respond to the warning preventively with a physical inspection. The system repeatedly measures after 2 minutes and comes to a decision point whether all 3 values (VOC, H2, temperature) have increased by more than 2% from the basic values, if not, the cycle returns to 5-minute VOC measurement intervals.

[0068] If this increase was above 2% from the basic values, a possible danger warning is sent again and after 2 minutes all three values are measured again. This type of warning follows on from the previous one, as the values are still increasing, even after 2 minutes from the previous measurement and have increased by at least another percentage compared to the previous decision block. This warning already requires the operator's (or airport employee's) attention, as these values are still increasing and it is possible that there is already a thermal runaway, perhaps even a starting fire, smoldering, or smoke. Then the employee can perform, for example, a visual inspection of the cargo or the container in which it is installed.

[0069] After 2 minutes from the measurement, it is checked whether the values have increased by 3% or more from the reference values. If not, the program returns to scanning VOCs in 5-minute intervals. If so, a danger warning is sent (it is no longer a possible danger, but a present danger). In order for this decision block to occur, the two previous decision blocks for the increase of three quantities by 1% or more, 2% or more had to be executed with a continuation to this block, i.e. a decision for YES (increase). Within 6 minutes, the current measured values increased by at least 3% or more compared to the basic values, if all three of these decision logics were fulfilled.

[0070] Further measurements of the quantities are performed (the increase is no longer examined, but only the current values, which are stored in the system and can thus be analyzed later) in short cycles and repeated warnings about this danger are sent every minute (infinitely), from this point the program never returns to VOC scanning in 5-minute intervals.

[0071] If the hydrogen concentration subsequently increases above 10% compared to the baseline value, warnings about impending danger are already sent (in this case, it is very likely that with such a concentration, a fire must already be present in order for the hydrogen concentration to increase, which is released from the battery, since the hydrogen concentration increases, there is a reasonable danger in the form of a possible explosion. The device will send these warnings indefinitely until it burns down or is destroyed by water during extinguishing.

[0072] In order for the device to return to VOC scanning in 5-minute intervals, the device would have to survive the fire and water and at the same time it would be necessary to disconnect and reconnect the power source so that the device starts a "clean" reset. Given the increase in the given measurements by more than 3% and subsequently possibly also hydrogen, it is very unlikely that it would still be in a state capable of operation.

[0073] The application, which serves to inform the user about measured values and warnings, also records the given warning historically. The operator can also see the number of warnings, e.g. for exceeding the values by 1 % (non-critical warning) during the day, week for the target air cargo container that he monitors. But at the same time, the system also notifies him immediately if a new warning occurs.

[0074] Industrial Applicability

[0075] The system and method for early detection of thermal runaway from a battery according to the invention can be used for checking aircraft components in air transport containing lithium batteries, which are pallets and containers including fireproof containers, aircraft, in particular the aircraft cabin, airport. Within the airport, security check, ground handling before departure, airport equipment intended for tracking the position in avionics, anchors, counters and systems, avionics systems for ground handling.

[0076] In addition to the above use, the system for early detection of thermal runaway from a battery according to the invention can also be used in other sectors of the economy - in healthcare and hospitals, in automobile and railway transport, in shipping and maritime transport, in business buildings and offices, in the hotel industry and in tourism and in households.

[0077] List of related tags

[0078] 1 - air cargo with detected battery

[0079] 11 - detected battery

[0080] 2 - gas concentration sensor

[0081] 3 - temperature sensor

[0082] 4 - microcontroller unit

[0083] 5 - wireless communication

[0084] 51 - LoRaWAN gateway (anchor)

[0085] 52 - BLE (Bluetooth Low Energy)

[0086] 53 - data transmission with Cat 1 bis technology (mobile operator antenna)

[0087] 6 - alarm

[0088] 61 - data visualization, user notification (computer)

[0089] 62 - data visualization, user notification (smartphone)

[0090] 7 - cloud storage with database

[0091] 8 - communication via satellite system

Claims

PATENT CLAIMS1. The early detection system of thermal runaway from a battery, containing an air cargo(I) with a detected battery (11), a gas concentration sensor (2) and / or a temperature sensor (3), microcontroller unit (4) with an advanced machine learning algorithm, wireless communication (5), an alarm (6) and a cloud storage (7) with a database, using Internet of Things technology, characterized in that the microcontroller unit (4) with an advanced machine learning algorithm, located within the range of the detected battery(II) air cargo (1) is connected to at least one gas concentration sensor (2) and / or at least one temperature sensor (3), wherein the microcontroller unit (4) with an advanced machine learning algorithm is connected by wireless communication (5) to the alarm (6) and a cloud storage (7) with a database.

2. The early detection system of thermal runaway from a battery according to claim 1 , characterized in that the gas concentration sensor (2) is configured to detect volatile organic compounds and hydrogen fluoride.

3. The early detection system of thermal runaway from a battery according to claim 1 , characterized in that the gas concentration sensor (2) is configured to detect volatile organic compounds and carbon monoxide.

4. The early detection system of thermal runaway from a battery according to claim 1 , characterized in that the gas concentration sensor (2) is configured to detect volatile organic compounds and hydrocarbons.

5. The early detection system of thermal runaway from a battery according to claim 1 , characterized in that the gas concentration sensor (2) is configured to detect volatile organic compounds and ethylene.

6. The early detection system of thermal runaway from a battery according to claim 1 , characterized in that the gas concentration sensor (2) is configured to detect volatile organic compounds and molecular hydrogen.

7. The early detection system of thermal runaway from a battery according to claim 1 , characterized in that the gas concentration sensor (2) is configured to detect hydrogen fluoride, carbon monoxide and carbon dioxide.

8. The early detection system of thermal runaway from a battery according to claim 1 , characterized in that the gas concentration sensor (2) is configured to detect molecular hydrogen, carbon monoxide and carbon dioxide.

9. The early detection system of thermal runaway from a battery according to claim 1 , characterized in that the gas concentration sensor (2) is configured to detect hydrocarbons, carbon monoxide and carbon dioxide.

10. The early detection system of thermal runaway from a battery according to claim 1, characterized in that the gas concentration sensor (2) is configured to detect ethylene, carbon monoxide and carbon dioxide.

11. The early detection system of thermal runaway from a battery according to claim 1 , characterized in that the gas concentration sensor (2) is configured to detect volatile organic compounds, carbon monoxide and carbon dioxide.

12. The early detection system of thermal runaway from a battery according to claim 1 , characterized in that temperature sensor (3) is a high precision temperature sensor with a sensing accuracy of ±1.5%, or an advanced temperature sensor with an accuracy of ± 0,5 % sensing accuracy.

13. The method of early detection of thermal runaway from the battery using the system according to any of claims 1 to 12, utilizing Internet of Things technology, characterized in that, information flows with data on gas concentration values and / or temperature values, recorded by gas concentration sensors (2) and / or temperature sensors (3), evaluated by a microcontroller unit (4) with an advanced machine learning algorithm, are sent by wireless communication (5) to an alarm (6) and to a cloud storage (7) with a database.