A method and system for thermal runaway management of energy storage batteries

Abstract

This invention provides a method for thermal runaway management of energy storage batteries, comprising: collecting cell temperature data, smoke data, and characteristic gas data of the energy storage battery, and performing data preprocessing; dynamically calculating anomaly judgment thresholds based on the preprocessed cell temperature data and in conjunction with the cell's historical temperature baseline, cycle count, and ambient temperature, thereby marking initial anomaly areas; converting the abnormal pixel coordinates of the initial anomaly areas into the three-dimensional physical coordinates and numbers of the cells through coordinate mapping; and determining the anomaly level by combining the preprocessed cell temperature data, smoke data, and characteristic gas data through a multimodal data fusion algorithm; performing graded blocking operations according to the anomaly level; generating alarm information and sending it to an external terminal, updating the anomaly level, and recording the handling data. The method and system of this invention can achieve cell-level anomaly location, graded blocking, and fire suppression, improving the operational efficiency and safety of battery swapping stations.

Description

Technical Field

[0001] This invention relates to the field of energy storage battery safety management technology, specifically to a method and system for thermal runaway management of energy storage batteries. Background Technology

[0002] With the rapid development of the new energy industry, energy storage batteries, as core components for energy storage and supply, have been widely used in battery swapping stations, energy storage power stations, electric vehicles, and other fields. However, under charge-discharge cycles, long-term use, or abnormal operating conditions, such as overcharging, short circuits, and high-temperature environments, energy storage batteries are prone to overheating, smoke, and even thermal runaway. Improper handling may lead to the spread of fire, causing equipment damage, battery swapping station shutdowns, and even endangering personnel safety. Therefore, thermal runaway management of energy storage batteries has become a core issue of concern in the industry.

[0003] In existing technologies, management solutions for thermal runaway in energy storage batteries mostly employ a single approach: temperature threshold-triggered alarms and fan cooling. However, this approach has two major drawbacks in practical applications:

[0004] I. Low anomaly location accuracy and poor maintenance efficiency. Existing solutions can only detect the overall temperature of the battery pack or battery cluster through temperature sensors, triggering an alarm when the temperature exceeds a preset threshold. However, they cannot distinguish whether the anomaly is caused by a single cell within the battery pack overheating, a localized cluster of cells overheating, or the entire battery pack overheating. Maintenance personnel must disassemble the battery pack one by one and check the temperature and status of each cell, resulting in long troubleshooting times and extended downtime at the battery swapping station, severely impacting operational efficiency. Furthermore, while some solutions incorporate image acquisition equipment, such as visible light cameras, to assist in positioning, they are limited by lighting interference within the battery compartment, such as shadows, reflections, and image resolution. They can only identify obvious anomalies such as smoke or fire, failing to achieve precise cell-level positioning and still falling short of the requirements for rapid maintenance.

[0005] Second, the existing solutions rely on a single approach, resulting in low efficiency and a lack of coordination in preventing thermal runaway. When facing the risk of thermal runaway, current solutions often depend solely on cooling with fans, failing to match differentiated response strategies based on the severity of the anomaly. For example, in the early stages of thermal runaway, such as smoke from the battery cell or a sudden temperature rise without fire, the extremely low efficiency of fan cooling makes it difficult for cold air to quickly cover the abnormal battery cell area, easily missing the optimal intervention time. In severe runaway scenarios where a fire has already started, cooling alone cannot suppress the spread of the fire. Furthermore, while some existing solutions include fire extinguishing devices, they often fail to follow up with timely cooling after extinguishing the fire, leaving residual heat that can easily cause the battery cell to reignite. In addition, existing solutions lack a multi-dimensional data fusion and judgment mechanism, relying solely on temperature thresholds to trigger alarms. This is prone to false alarms due to environmental temperature differences, such as localized high temperatures inside the battery compartment in summer, and sensor errors, leading to unnecessary downtime losses.

[0006] Therefore, there is an urgent need for a solution that can overcome the shortcomings of existing technologies in the management of thermal runaway of energy storage batteries, such as low positioning accuracy, single handling methods, poor coordination, and high false alarm rate, to achieve cell-level anomaly positioning, graded and precise blocking and fire extinguishing, and improve the operational efficiency and safety of battery swapping stations. Summary of the Invention

[0007] Based on the technical problems described above, this invention provides a method and system for thermal runaway management of energy storage batteries. Through the collaborative design of precise sensing, graded blocking, intelligent control and external linkage, it achieves cell-level positioning, graded disposal and collaborative control, thus overcoming the deficiencies of existing technologies.

[0008] Specifically, according to one aspect of the present invention, a method for thermal runaway management of an energy storage battery is provided, the method comprising the following steps:

[0009] Step S1: Collect cell temperature data, smoke data, and characteristic gas data of the energy storage battery, and perform data preprocessing;

[0010] Step S2: Mark the initial abnormal area based on the preprocessed cell temperature data, dynamically calculate the abnormal judgment threshold by combining the cell historical temperature baseline, cycle number and ambient temperature, and convert the abnormal pixel coordinates into the three-dimensional physical coordinates and number of the cell by coordinate mapping. Combine the preprocessed cell temperature data, smoke data and characteristic gas data, and determine the abnormal level by multimodal data fusion algorithm.

[0011] Step S3: Perform graded blocking operations according to the abnormality level;

[0012] Step S4: Generate alarm information and send it to the external terminal, update the anomaly level and record the handling data.

[0013] According to certain preferred embodiments of the present invention, in step S1, the data preprocessing includes: applying a moving average filter to the cell temperature data, applying a threshold filter to the smoke data, applying a Kalman filter to the characteristic gas data; automatically calibrating sensor data every 24-72 hours, and determining sensor fault and removing faulty data when abnormalities are detected 3-5 times consecutively.

[0014] According to certain preferred embodiments of the present invention, in step S2, the dynamic calculation of the anomaly determination threshold by combining the historical temperature baseline of the battery cell, the number of cycles, and the ambient temperature includes:

[0015] A baseline threshold is determined from the historical temperature baseline of the battery cell;

[0016] The base threshold is adjusted based on the number of battery cell cycles and the ambient temperature, wherein:

[0017] When the number of cycles is ≤500, the threshold increases by 15℃; when 500 < number of cycles ≤1000, the threshold increases by 10℃; when the number of cycles >1000, the threshold increases by 5℃; when the ambient temperature >35℃, the threshold decreases by 2℃; when the ambient temperature <5℃, the threshold increases by 2℃.

[0018] According to certain preferred embodiments of the present invention, in step S2, the conversion of abnormal pixel coordinates into three-dimensional physical coordinates and numbers of the battery cell through coordinate mapping includes: during the system initialization phase, obtaining the reference three-dimensional coordinates and dimensions of the battery pack through a laser rangefinder and calibration tools, calculating the theoretical three-dimensional coordinates of each battery cell according to the cell arrangement model, and establishing a mapping table between the cell coordinates and the pixel coordinates of the infrared image; after identifying the abnormal area, querying the mapping table to convert the abnormal pixel coordinates into three-dimensional physical coordinates.

[0019] According to certain preferred embodiments of the present invention, in step S2, the determination of the anomaly level by combining the preprocessed cell temperature data, smoke data, and characteristic gas data using a multimodal data fusion algorithm includes:

[0020] When the temperature is ≥60℃ and the smoke concentration is <0.1mg / m³ 3 When CO concentration is <10ppm, H2S concentration is <5ppm, and SO2 concentration is <2ppm, it is judged as ordinary superheating.

[0021] When the temperature is ≥60℃ and the smoke concentration is ≥0.1mg / m³ 3 A thermal runaway is defined as the initial stage when at least one of the following conditions is met: CO concentration ≥ 10 ppm, H2S concentration ≥ 5 ppm, or SO2 concentration ≥ 2 ppm.

[0022] When the temperature is ≥100℃ and the smoke concentration is ≥0.5mg / m³ 3 A situation is considered severely out of control if at least one of the following conditions is met: CO concentration ≥ 50 ppm; or the presence of an area with a temperature ≥ 300°C.

[0023] According to certain preferred embodiments of the present invention, in step S3, the graded blocking operation includes:

[0024] If it is ordinary overheating, start the corresponding honeycomb-type zoned air duct and directional air cooler to deliver airflow for cooling;

[0025] If it is the initial stage of thermal runaway, start directional air supply for cooling at maximum power. If the situation escalates within a set time, start aerosol fire suppression and send a command to the circuit cut-off unit to cut off the charging and / or discharging circuit of the abnormal battery pack to remove the continuous heat source.

[0026] In case of severe loss of control, activate aerosol fire suppression, close fire doors, activate fire sprinklers, and disconnect the abnormal and adjacent battery pack circuits.

[0027] According to certain preferred embodiments of the present invention, in step S4, the alarm information includes the abnormal cell number, location, handling status, and suggested measures, and the external terminals include the battery swapping station monitoring center and the fire department.

[0028] According to certain preferred embodiments of the present invention, the method further includes a false alarm suppression mechanism:

[0029] No alarm is triggered when a single sensor shows abnormal data.

[0030] When compared with historical data from the same period over the past 7 days, a deviation of less than 10% is considered normal fluctuation.

[0031] According to another aspect of the present invention, a thermal runaway management system for an energy storage battery is provided, the thermal runaway management system comprising:

[0032] The precision sensing module is used to collect cell temperature data, smoke data, and characteristic gas data of the energy storage battery to achieve cell-level anomaly location.

[0033] The intelligent control module is used to determine the anomaly level through a multimodal data fusion algorithm and output control commands;

[0034] The graded blocking module is used to perform differentiated handling operations based on the level of abnormality.

[0035] The external linkage module is used to send alarm information to external terminals and receive feedback.

[0036] According to certain preferred embodiments of the present invention, the precise sensing module includes:

[0037] The infrared thermal imager unit is used to collect the temperature of the battery cells of the energy storage battery and image them, and dynamically adjust the resolution according to the laser ranging results, and identify abnormal cells through dynamic threshold and edge detection.

[0038] Multi-sensor array units, including smoke sensors and characteristic gas sensors, are deployed on the top of the battery compartment, in the gaps between battery packs, or in key ventilation paths;

[0039] The coordinate mapping unit is used to establish the mapping relationship between the physical coordinates and pixel coordinates of the battery cell, and output the number and three-dimensional coordinates of the abnormal battery cell.

[0040] According to certain preferred embodiments of the present invention, the infrared thermal imager unit dynamically adjusts its resolution based on the detection distance:

[0041] When the detection distance is greater than 1.5m, the resolution is increased to 640×512 pixels;

[0042] When the detection distance is ≤1.5m, the resolution is reduced to 320×240 pixels and Gaussian filtering is performed.

[0043] According to certain preferred embodiments of the present invention, the intelligent control module includes:

[0044] A multimodal data fusion algorithm unit is used to determine the anomaly level and suppress false alarms;

[0045] The fire extinguishing-cooling coordinated timing control unit is used to control the graded blocking module to perform operations in a time sequence;

[0046] Anomaly level determination unit is used to update the anomaly level based on real-time data.

[0047] According to certain preferred embodiments of the present invention, the graded blocking module includes:

[0048] The honeycomb partitioned air duct unit adopts a matrix layout, with each air duct corresponding to 2-3 battery cells, and the air outlet angle is adjusted by an electromagnetic airflow guide valve.

[0049] The directional air cooler unit, linked with the honeycomb partitioned air duct unit, is used to provide cooling airflow;

[0050] The aerosol fire suppression unit is installed on top of the battery pack, and the spray nozzle angle is adjustable.

[0051] The circuit cutoff unit is linked with the battery management system (BMS) of the battery pack and supports command cutoff and independent hardware emergency cutoff.

[0052] Fire door control unit, used to close fire doors in case of severe loss of control;

[0053] The fire sprinkler unit is located on top of the battery compartment and is used for spraying water to cool down and extinguish fires.

[0054] According to certain preferred embodiments of the present invention, the circuit breaker unit is provided with an independent hardware safety link, which directly drives the circuit breaker to perform emergency disconnection when a sudden surge in open flame or smoke concentration is detected.

[0055] According to certain preferred embodiments of the present invention, the deployment density of the multi-sensor array unit is flexibly configured based on the battery compartment volume, ventilation conditions, and risk level.

[0056] According to certain preferred embodiments of the present invention, the external linkage module supports sending alarm information to the battery swapping station monitoring center and the fire department, and enables SMS alarm when communication fails.

[0057] According to certain preferred embodiments of the present invention, the system further includes a data calibration mechanism:

[0058] The sensor is automatically calibrated every 24-72 hours, and the parameters are automatically adjusted when the error exceeds 10%.

[0059] If an abnormality is detected 3-5 times consecutively, the sensor is deemed faulty and an alarm is triggered.

[0060] According to certain preferred embodiments of the present invention, the system supports closed-loop data management, recording the handling process and effect of each anomaly for subsequent algorithm optimization and system upgrades.

[0061] By employing the method and system of this invention, a solution can be achieved to realize cell-level anomaly location, graded and precise blocking and fire extinguishing, thereby improving the operational efficiency and safety of battery swapping stations. Attached Figure Description

[0062] The accompanying drawings are provided in this specification to more clearly explain the technical solutions of the present invention; however, the art is not limited thereto. The flowcharts shown in the drawings are merely illustrative and do not necessarily include all contents and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be decomposed, combined, or partially merged, so the actual execution order may change according to the actual situation.

[0063] Figure 1 This is a flowchart illustrating a method for thermal runaway management of an energy storage battery according to one embodiment of the present invention.

[0064] Figure 2 This is a schematic diagram of a system architecture for thermal runaway management of an energy storage battery according to one embodiment of the present invention;

[0065] Figure 3 This is a schematic diagram of the structure of a precision sensing module according to one embodiment of the present invention;

[0066] Figure 4 This is a schematic diagram of the main working steps of the linkage between steps S2 and S3 according to an embodiment of the present invention;

[0067] Figure 5 This is a schematic diagram of the structure of a graded blocking module according to one embodiment of the present invention.

[0068] Explanation of reference numerals in the attached figures:

[0069] S1-S4. Steps and flow of thermal runaway management methods for energy storage batteries; 10. Thermal runaway management system; 1. Precision sensing module; 11. Infrared thermal imager unit; 12. Multi-sensor array unit; 13. Coordinate mapping unit; 2. Intelligent control module; 21. Multimodal data fusion algorithm unit; 22. Fire extinguishing-cooling coordinated timing control unit; 23. Anomaly level determination unit; 3. Graded blocking module; 31. Honeycomb zoned air duct unit; 32. Directional air cooler unit; 33. Aerosol fire extinguishing unit; 34. Circuit cut-off unit; 35. Fire door control unit; 36. Fire sprinkler unit; 4. External linkage module. Detailed Implementation

[0070] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0071] Example 1

[0072] like Figure 1 and Figure 4 As shown below, the method flow for thermal runaway management of the energy storage battery of the present invention is described in detail below, and the specific scheme is as follows:

[0073] Step S1: Collect data on cell temperature, smoke, and characteristic gases from the energy storage battery, and perform data preprocessing.

[0074] The infrared thermal imager unit 11 collects the cell temperature of the energy storage battery and images the data. Its built-in laser ranging module is used to dynamically adjust the resolution. The multi-sensor array unit 12 collects the smoke concentration in the battery compartment and the concentration of gases such as CO, H2S, and SO2. The sensor data is automatically calibrated every 24-72 hours, and faulty data is removed. The collected cell temperature data is filtered by moving average, the smoke data is filtered by threshold, and the gas data is filtered by Kalman to eliminate noise interference.

[0075] Step S2: Cell-level anomaly location and anomaly level determination

[0076] For the cell temperature data processed in step S1, the initial abnormal area is marked. The abnormality judgment threshold is dynamically calculated by combining the cell's historical temperature baseline, the number of cycles, and the ambient temperature. The abnormal pixel coordinates are converted into the three-dimensional physical coordinates and number of the cell through coordinate mapping. Combined with the preprocessed cell temperature, smoke, and characteristic gas data, the abnormality level is determined by a multimodal data fusion algorithm. For example, there are three levels: ordinary overheating, early stage of thermal runaway, and severe runaway. The system also checks whether there are false alarms.

[0077] Step S3: Perform graded blocking operations based on the anomaly level.

[0078] Based on the anomaly level determined in step S2, differentiated handling is performed, and the specific differentiated handling is as follows:

[0079] S3.1: If it is determined to be ordinary overheating, then according to the coordinates of the abnormal battery cell, determine the corresponding honeycomb partition air duct unit 31 number, control the electromagnetic airflow guide valve to rotate to the angle aligned with the abnormal battery cell, and start the directional air cooler to perform air supply and cooling; at the same time, monitor the air duct pressure in real time and adjust the air outlet angle as appropriate; when the temperature of the abnormal battery cell detected in real time is lower than a certain threshold, close the air duct outlet and the air cooler.

[0080] S3.2: If the situation is determined to be in the early stage of thermal runaway, the highest priority blocking measures will be activated first: based on the coordinates of the abnormal battery cell, the corresponding honeycomb partition air duct and directional cooling fan will be controlled to deliver directional airflow to the abnormal battery cell at maximum power to attempt to suppress the thermal runaway trend in the early stage; at the same time, a command will be sent to the circuit cutoff unit 34 to cut off the charging and / or discharging circuit of the abnormal battery pack to remove the continuous heat source. The system monitors the handling effect in real time. If the temperature continues to rise or the smoke and / or gas concentration increases within a set time, the situation will be determined to be escalated, and the system will immediately switch to the severe runaway handling process with fire extinguishing function as described in S3.3.

[0081] S3.3: If the situation is determined to be seriously out of control, the aerosol fire extinguishing unit 33 will be activated to suppress the abnormal area according to the coordinated timing rules. Simultaneously, the fire door will be closed to isolate the area, the fire sprinkler unit 36 ​​will be activated to cool down and extinguish the fire, and the charging circuit of the abnormal battery pack and adjacent battery packs will be cut off.

[0082] Step S4: External Linkage and Alarms

[0083] An alarm message is generated, including basic information about the abnormal battery cell, its location, handling status, and suggested measures. This alarm message is then sent to external terminals, such as the battery swapping station monitoring center or the fire department. An SMS alarm is activated if communication fails. After the external terminal parses the alarm message, maintenance personnel perform repairs based on the abnormal battery cell's coordinates, and the fire department conducts emergency response according to the suggested measures.

[0084] Furthermore, the anomaly level is updated at a set frequency. If the anomaly level is downgraded, the corresponding handling process is stopped; if the anomaly level is upgraded, the handling process for the corresponding level is switched. After the handling is completed, the anomaly information, handling measures, and effects are recorded to form closed-loop data for subsequent algorithm optimization.

[0085] The method for managing thermal runaway in energy storage batteries proposed in this invention has the following advantages compared to existing technologies:

[0086] The thermal runaway management method of this invention can locate abnormal energy storage batteries at the cell level, improving maintenance efficiency; multimodal fusion detection data can greatly reduce false alarm rate and reduce unnecessary downtime losses; and in conjunction with a graded blocking mechanism, it can improve the blocking success rate and effectively prevent the spread of fire; when severely out of control, it can precisely link with external systems to improve fire fighting efficiency.

[0087] Example 2

[0088] like Figure 2-5 As shown, the energy storage battery thermal runaway management system 10 of the present invention specifically includes a precise sensing module 1, an intelligent control module 2, a graded blocking module 3, and an external linkage module 4. Each module has independent functions and works in synergy. The specific structure and functions are as follows:

[0089] 1. Precision Sensing Module 1

[0090] The precision sensing module 1 is used to collect data on cell temperature, smoke, and characteristic gases of energy storage batteries, and to achieve cell-level anomaly location by first marking the approximate area and then optimizing the edge accuracy. The precision sensing module 1 includes an infrared thermal imager unit 11, a multi-sensor array unit 12, and a coordinate mapping unit 13.

[0091] (1) Infrared thermal imager unit 11, wherein the infrared thermal imager unit 11 adopts a high-sensitivity infrared detector with a detection wavelength range of 8-14 sensitivity. The resolution can be dynamically adjusted according to the distance information obtained by the built-in laser ranging module. When the detection distance is >1.5m (long-distance imaging), the resolution is increased to 640 pixels. The edge of the battery cell is enhanced by gray value gradient calculation to improve the outline clarity. When the detection distance is ≤ outline clarity (near-distance imaging), the resolution is reduced to 320 pixels. High-frequency thermal noise is filtered by Gaussian filtering, so that the temperature detection error is controlled within ±3℃.

[0092] The initial abnormal area is marked in the following manner. Considering the individual differences of battery cells, such as aging degree, battery type and environmental variables, low-aging cells are prone to false alarms, such as normal temperature fluctuations exceeding a fixed threshold, or high-aging cells are missed, such as due to local overheating not reaching the fixed threshold. Therefore, after acquiring the cell temperature and imaging it with an infrared thermal imager, this invention calls up the basic parameters of each cell (the basic parameters include: cell type, number of cycles and temperature tolerance threshold at the time of manufacture), and retrieves the historical temperature data of the cell over the past 72 hours to establish a historical temperature baseline (T-base) for a single cell, thereby determining the basic threshold. The basic threshold is then dynamically adjusted based on its number of cycles and ambient temperature to determine the abnormal judgment threshold (T-th). Specifically:

[0093] When the number of battery cell cycles is ≤500 (belonging to a low aging level), then: T-th = T-base + 15℃;

[0094] When 500 cycles < cell cycle count ≤ 1000 cycles (belonging to the medium aging level), then: T-th = T-base + 10℃;

[0095] When the number of battery cell cycles exceeds 1000 (which is considered a high degree of aging), then: T-th = T-base + 5℃;

[0096] The temperature fluctuation tolerance varies depending on the degree of cell aging, and therefore the threshold rises by different amounts.

[0097] Meanwhile, if the current ambient temperature is >35℃ (high-temperature summer conditions), the threshold is further reduced by 2℃ based on the above calculation results; if the ambient temperature is <5℃ (low-temperature winter conditions), the threshold is further increased by 2℃. Only cells with a temperature ≥T-th are marked as the initial abnormal area, which ensures that the threshold is adapted to the impact of ambient temperature on cell heat dissipation.

[0098] Then, considering that edge detection of the battery cell is incomplete during long-distance imaging and that false edges increase during close-range imaging, edge detection of the initial abnormal area is performed by combining the Canny operator with the distance parameters obtained from the infrared thermal imager. The specific method is as follows:

[0099] The infrared thermal imager acquires the real-time distance D between itself and the target battery pack through its built-in laser ranging module. Based on the distance D, it classifies the detection scale level and matches the corresponding Canny operator parameters, including template size and threshold. Canny edge detection is performed according to the matched parameters. For long-distance imaging, Gaussian filtering preprocessing is performed first, and for close-distance imaging, short edge removal postprocessing is performed after edge detection. The standard size of the battery cell is retrieved through the coordinate mapping unit 13. The detected battery cell edge contour size is compared with the standard size. If the deviation is >10%, the Canny threshold of the current scale is automatically fine-tuned, and the detection is repeated until the deviation is ≤10%.

[0100] Finally, by combining the standard dimensions of the battery cells, the number and location of the battery cells contained in the abnormal area are identified.

[0101] In addition to the infrared thermal imager unit 11 for temperature detection, the system of the present invention also collects smoke data and characteristic gas data through the multi-sensor array unit 12, which can capture risk signals such as smoke and characteristic gases earlier, solving the problem that relying solely on temperature detection may result in late detection and missing the best time for handling.

[0102] (2) Multi-sensor array unit 12

[0103] The multi-sensor array unit 12 includes a smoke sensor deployed on top of the battery compartment, and gas sensors (using electrochemical principles to detect gases such as CO, H2S, and SO2) deployed in the gaps between the battery packs, on top, or in key ventilation paths. The deployment density can be flexibly configured according to the battery compartment volume, ventilation conditions, and risk level to ensure detection sensitivity. The multi-sensor array unit 12 automatically performs data calibration every 72 hours; if the error is greater than 10%, it automatically adjusts the parameters; and if three consecutive abnormal detections are detected, a sensor malfunction is determined and an alarm is triggered.

[0104] (3) Coordinate mapping unit 13

[0105] During the system initialization and installation phase, a laser rangefinder and calibration tools are used to accurately acquire the baseline three-dimensional physical coordinates and spatial dimensions of each battery pack within the battery compartment. The coordinate mapping unit 13 automatically calculates the theoretical three-dimensional coordinates of each cell in physical space based on a pre-entered cell arrangement model within the battery pack, and establishes a mapping table between these coordinates and the pixel coordinates in the images acquired by the infrared thermal imager unit 11. When an abnormal area is identified, this mapping table is consulted to convert the abnormal pixel coordinates into the precise physical location and number of the cell, which is then output.

[0106] The precise sensing module 1 can optimize the imaging parameters of the infrared thermal imager, ensuring that clear and low-noise infrared images of the battery cells can be obtained at different distances. Furthermore, abnormal battery cell areas can be accurately marked from the optimized images to determine the number and physical location of abnormal battery cells.

[0107] 2. Intelligent Control Module 2

[0108] The intelligent control module 2 is the core decision-making unit of the system of the present invention, including: a multimodal data fusion algorithm unit 21, a fire extinguishing-cooling coordinated timing control unit 22, and an anomaly level determination unit 23, to realize data processing, timing control and level update.

[0109] (1) Multimodal data fusion algorithm unit 21

[0110] The multimodal data fusion algorithm unit 21 preprocesses and fuses the acquired temperature data, smoke data, and characteristic gas data. For example, when the temperature is ≥60℃ and the smoke concentration is <0.1mg / m³, the fusion is determined accordingly. 3 When CO concentration is <10ppm, H2S concentration is <5ppm, and SO2 concentration is <2ppm, it is judged as ordinary superheating; when the temperature is ≥60℃ and the smoke concentration is ≥0.1mg / m³, it is considered ordinary superheating. 3 Thermal runaway is defined as the initial stage when CO concentration is ≥10 ppm and / or H2S concentration is ≥5 ppm and / or SO2 concentration is ≥2 ppm; when the temperature is ≥100℃ and the smoke concentration is ≥0.5 mg / m³.3 A situation is considered severely out of control if the CO concentration is ≥50ppm and / or if there is a temperature ≥300℃.

[0111] The multimodal data fusion algorithm unit 21 is also equipped with a false alarm suppression mechanism, such as: no alarm when a single dimension triggers the threshold; and when compared with historical data of the same period in the past 7 days, a deviation of <10% is judged as a normal fluctuation range.

[0112] (2) Fire extinguishing-cooling coordinated timing control unit 22

[0113] The fire extinguishing-cooling coordinated timing control unit 22 is used to issue coordinated timing commands to the following graded blocking modules 3 according to the anomaly level and the treatment effect, such as:

[0114] If it is a normal overheating, according to the coordinates of the abnormal cell, the corresponding honeycomb partition air duct unit 31 and directional air cooler unit 32 are instructed to start and perform directional air supply cooling at normal power.

[0115] In the early stages of thermal runaway, the graded blocking module 3 is given priority to execute directional cooling at maximum power. If cooling is ineffective and the situation escalates, aerosol fire suppression is immediately initiated. After the open flame is suppressed, the duct unit is activated again for cooling to remove residual heat and completely block the risk of reignition. The fire suppression-cooling coordinated timing control unit 22 can also dynamically adjust parameters according to the ambient temperature, such as increasing the cooling airflow when the ambient temperature is >35℃.

[0116] In case of severe loss of control, activate aerosol fire suppression, close fire doors, activate fire sprinklers, and disconnect the abnormal and adjacent battery pack circuits.

[0117] (3) Abnormal level determination unit 23

[0118] The anomaly level determination unit 23 is used to update the thermal runaway level and determine the level of the real-time situation based on the above conditions of ordinary overheating, initial stage of thermal runaway, and severe runaway. Furthermore, the level is not adjusted when a single sensor data is abnormal, but the corresponding level is determined when two or more dimensions of data are abnormal, ensuring the reliability of the determination.

[0119] 3. Graded blocking module 3

[0120] The graded blocking module 3 performs differentiated handling operations according to the level of abnormality, including: honeycomb partitioned air duct unit 31, directional air cooler unit 32, aerosol fire extinguishing unit 33, circuit cut-off unit 34, fire door control unit 35, and fire sprinkler unit 36. Each unit works in coordination according to the instructions of the above-mentioned intelligent control module 2.

[0121] (1) Honeycomb partitioned air duct unit 31

[0122] This invention employs a matrix layout of honeycomb partitioned air duct units 31 covering the battery compartment. Each air duct unit is a cylindrical structure with a diameter of 50mm, corresponding to 2-3 battery cells. Directional air outlets are opened on the side walls, and the angle can be adjusted by an electromagnetic airflow guide valve. An airflow equalization net is provided at the air outlet. The honeycomb partitioned air duct unit 31 determines the air duct unit number according to the coordinates of the abnormal battery cell. For example, "B3 pack, 5th row" corresponds to air duct B3-5. The electromagnetic airflow guide valve is controlled to rotate to the angle aligned with the abnormal battery cell.

[0123] (2) Directional air cooler unit 32

[0124] The directional air cooler unit 32 is linked to the pressure sensor of the air duct unit to provide cooling airflow.

[0125] (3) Aerosol fire extinguishing unit 33

[0126] The aerosol extinguishing unit 33 is installed at the top center of the battery pack. Each aerosol extinguishing unit 33 corresponds to one battery pack. The spray nozzle is equipped with an adjustable angle nozzle to spray the agent according to the location of the abnormal battery cell.

[0127] (4) Circuit disconnection unit 34

[0128] The circuit disconnection unit 34 is linked to the battery management system (BMS) of the battery pack and executes the disconnection operation according to the instructions of the intelligent control module 2. When an extreme danger signal such as a momentary open flame signal or a rapid increase in smoke concentration is detected, the main control logic will be bypassed and the circuit breaker will be directly driven to perform a millisecond-level emergency disconnection.

[0129] (5) Fire door control unit 35

[0130] The fire door control unit 35 is used to drive the door to close and block smoke and flames in the event of severe loss of control.

[0131] (6) Fire sprinkler unit 36

[0132] The fire sprinkler unit 36 ​​is located on the top of the battery compartment. It is used to open the sprinkler head when there is a serious loss of control or when the temperature exceeds a certain value after the fire door is closed, and to open the drain valve according to the degree of water accumulation to avoid soaking the normal battery cells.

[0133] 4. External linkage module 4

[0134] The external linkage module 4 is used to send alarm information (including abnormal cell number, location, and handling suggestions) to external terminals such as the battery swapping station monitoring center and fire department, and automatically activate SMS alarm when communication fails to ensure that the information is delivered.

[0135] The thermal runaway management system 10 for energy storage batteries proposed in this invention, compared with the prior art, achieves accurate detection of abnormal cells in a short time through a precise sensing module, solving the problem of inefficient maintenance caused by ambiguous positioning; relying on the three-stage coordinated timing of the graded blocking module and the intelligent control module, it matches differentiated handling strategies for ordinary overheating, the initial stage of thermal runaway, and severe runaway, which can effectively improve the success rate of blocking in the initial stage of thermal runaway and reduce the reignition rate, solving the defects of single handling and high reignition rate in the prior art, and significantly improving the operation efficiency of battery swapping stations and the safety management level of energy storage batteries.

[0136] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus or device (such as a computer-based system, a processor-included system or other system that can fetch and execute instructions from, an instruction execution system, apparatus or device).

[0137] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0138] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0139] It should also be further understood that the term "and / or" as used in this specification refers to any combination of one or more of the associated listed items, as well as all possible combinations, and includes such combinations.

[0140] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for thermal runaway management of an energy storage battery, the method comprising: The method comprises the following steps: Step S1: collecting cell temperature data, smoke data and characteristic gas data of the energy storage battery, and performing data preprocessing; Step S2: dynamically calculating an abnormality determination threshold based on the preprocessed cell temperature data in combination with a historical temperature baseline of the cell, a cycle number and an ambient temperature, thereby marking an initial abnormal area, and converting abnormal pixel coordinates of the initial abnormal area into three-dimensional physical coordinates and numbers of the cell through coordinate mapping, and determining an abnormality level through a multi-modal data fusion algorithm in combination with the preprocessed cell temperature data, smoke data and characteristic gas data; Step S3: performing a hierarchical blocking operation according to the abnormality level; Step S4: generating alarm information and sending it to an external terminal, updating the abnormality level and recording disposal data, wherein: In the step S2, the dynamically calculating of the abnormality determination threshold in combination with the historical temperature baseline of the cell, the cycle number and the ambient temperature comprises: determining a basic threshold from the historical temperature baseline of the cell; adjusting the basic threshold according to the cycle number of the cell and the ambient temperature, wherein: when the cycle number is less than or equal to 500 times, the threshold is increased by 15℃; when 500 times < cycle number < 1000 times, the threshold is increased by 10℃; when the cycle number is greater than 1000 times, the threshold is increased by 5℃; when the ambient temperature is greater than 35℃, the threshold is decreased by 2℃; when the ambient temperature is less than 5℃, the threshold is increased by 2℃, and In the step S3, the performing of the hierarchical blocking operation according to the abnormality level comprises: if it is a general heat, starting directional air supply cooling of the corresponding honeycomb type partitioned air duct and directional air cooler; if it is a heat runaway initial stage, starting directional air supply cooling with maximum power, if the situation is upgraded within the set time, starting aerosol fire extinguishing, and sending an instruction to the loop cutting unit to cut off the charging and / or discharging loop of the abnormal battery pack to remove the continuous heat source; if it is a serious runaway, starting aerosol fire extinguishing, closing the fireproof door, starting fire sprinkling, and cutting off the loop of the abnormal and adjacent battery packs.

2. The method of claim 1, wherein, In the step S1, the data preprocessing comprises: applying sliding average filtering to the cell temperature data, threshold filtering to the smoke data, and Kalman filtering to the characteristic gas data; automatically calibrating the sensor data every 24-72 hours, and determining sensor failure and removing faulty data when abnormality is detected for 3-5 times continuously.

3. The method of claim 1, wherein: In the step S2, the converting of the abnormal pixel coordinates into three-dimensional physical coordinates of the cell through coordinate mapping comprises: In the system initialization stage, acquiring the reference three-dimensional coordinates and dimensions of the battery pack through a laser range finder and a calibration tool, calculating the theoretical three-dimensional coordinates of each cell according to a cell arrangement model, and establishing a mapping table of the theoretical three-dimensional coordinates and infrared image pixel coordinates; after identifying the abnormal area, querying the mapping table to convert the abnormal pixel coordinates into three-dimensional physical coordinates.

4. The method of claim 1, wherein: In the step S2, the determining of the abnormality level through a multi-modal data fusion algorithm in combination with the preprocessed cell temperature data, smoke data and characteristic gas data comprises: When the temperature is ≥ 60 °C and the smoke concentration < 0.1 mg / m 3 , the CO concentration < 10 ppm, the H2S concentration < 5 ppm, and the SO2 concentration < 2 ppm, the determination is that the general heat is passed. When the temperature is ≥ 60°C and at least one of the following is satisfied: smoke ≥ concentration 0.1 mg / m 3 CO concentration ≥ 10 ppm, H2S concentration ≥ 5 ppm, SO2 concentration ≥ 2 ppm, the thermal runaway incipient stage is determined. When at least one of the temperature ≥ 100℃ and the smoke concentration ≥ 0.5 mg / m 3 , the CO concentration ≥ 50 ppm, and the presence of a region with a temperature ≥ 300℃ is satisfied, it is determined that the serious runaway has occurred.

5. A thermal runaway management system for use in a thermal runaway management method of the energy storage battery according to any one of claims 1-4, characterized in that, Comprise: Precise perception module for collecting cell temperature data, smoke data and characteristic gas data of the energy storage battery, realizing cell-level abnormal positioning; Intelligent control module for determining abnormal level through multi-modal data fusion algorithm and outputting control instructions; Hierarchical blocking module for performing differentiated disposal operation according to abnormal level; External linkage module for sending alarm information to external terminal and receiving feedback, wherein: The precise perception module comprises: Infrared thermal imager unit for collecting the cell temperature data of the energy storage battery and imaging, and dynamically adjusting the resolution according to the laser ranging result, identifying abnormal cells through dynamic threshold and edge detection; Multi-sensor array unit including smoke sensor and characteristic gas sensor, deployed on the top of the battery compartment, the gap between battery packs or the key ventilation path; Coordinate mapping unit for establishing the mapping relationship between the physical coordinates and pixel coordinates of the cells, outputting the number and three-dimensional coordinates of the abnormal cells, and The intelligent control module comprises: Multi-modal data fusion algorithm unit for determining abnormal level and suppressing false positives; Fire extinguishing-cooling coordination timing control unit for controlling the hierarchical blocking module to execute operation in time sequence; Abnormal level determination unit for updating abnormal level according to real-time data.

6. The system of claim 5, wherein, The hierarchical blocking module comprises: Honeycomb type partitioned air duct unit with matrix layout, each air duct corresponding to 2-3 cell groups, the outlet angle being adjusted by electromagnetic air flow guide valve; Directional air cooler unit linked with the honeycomb type partitioned air duct unit for providing cooling air flow; Aerosol fire extinguishing unit installed on the top of the battery pack, with adjustable jet angle; Loop cutting unit linked with the battery management system of the battery pack, supporting command cutting and independent hardware emergency cutting; Fireproof door control unit for closing the fireproof door in case of serious out-of-control; Fire sprinkler unit arranged on the top of the battery compartment for sprinkling cooling and fire extinguishing.

Images