Battery management system and battery thermal runaway early warning method

Abstract

The invention discloses a battery management system and a battery thermal runaway early warning method, and relates to the technical field of battery safety management, the battery management system comprises a temperature sensor and a controller, and the temperature sensor is in communication connection with the controller; the temperature sensor is used for acquiring temperature data of a vehicle battery in a current time period; the controller is used for determining thermal runaway modes according to the temperature data, wherein the thermal runaway modes comprise a mechanical runaway mode, an electric runaway mode and a high-temperature runaway mode; determining a target adjustment coefficient according to the thermal runaway mode, and determining a target thermal runaway threshold according to the target adjustment coefficient and a preset thermal runaway threshold; and under the condition that the temperature rise rate of the latter time period is greater than the target thermal runaway threshold value, an early warning instruction is generated, and the early warning instruction is used for indicating to send battery thermal runaway early warning to a user, so that the technical problem of how to accurately carry out thermal runaway early warning on the vehicle battery is solved, and the accuracy of judging the thermal runaway of the vehicle battery is improved.

Description

Technical Field

[0001] This application relates to the field of battery safety management technology, and more specifically, to a battery management system and a battery thermal runaway early warning method. Background Technology

[0002] With the widespread adoption of electric vehicles and energy storage power stations, the safety of lithium-ion batteries, as the core energy storage carrier, is becoming increasingly prominent. Thermal runaway, the most destructive failure mode of batteries, often results in smoke, fire, or even explosion, seriously threatening personal safety and public property. Currently, mainstream battery management systems typically use a threshold-based method based on the rate of temperature rise for thermal runaway monitoring. This method uses a single standard and cannot distinguish the characteristics of thermal runaway under different triggering conditions. The temperature evolution patterns of batteries differ significantly under different triggering modes, and using a uniform rate of temperature rise threshold can easily lead to misjudgments or missed detections, resulting in inaccurate warning times. Therefore, there is a technical challenge in related technologies to accurately provide thermal runaway warnings for vehicle batteries.

[0003] There is still no effective solution to the technical problem of how to accurately provide early warning of thermal runaway in vehicle batteries. Summary of the Invention

[0004] This application provides a battery management system and a battery thermal runaway early warning method to at least solve the technical problem of how to accurately provide thermal runaway early warning for vehicle batteries in related technologies.

[0005] According to one embodiment of this application, a battery management system is provided, including a temperature sensor and a controller, wherein the temperature sensor is communicatively connected to the controller; the temperature sensor is used to acquire temperature data of a vehicle battery in the current time period; the controller is used to: determine a thermal runaway mode based on the temperature data, the thermal runaway mode including a mechanical runaway mode, an electrical runaway mode, and a high-temperature runaway mode; determine a target adjustment coefficient based on the thermal runaway mode, and determine a target thermal runaway threshold based on the target adjustment coefficient and a preset thermal runaway threshold; and generate a warning command if the temperature rise rate in a later time period exceeds the target thermal runaway threshold, the warning command being used to instruct a battery thermal runaway warning to be issued to the user.

[0006] According to another embodiment of this application, a battery thermal runaway early warning method is provided, applied to a battery management system, comprising: acquiring temperature data of a vehicle battery in the current time period; determining a thermal runaway mode based on the temperature data, the thermal runaway mode including a mechanical runaway mode, an electrical runaway mode, and a high-temperature runaway mode; determining a target adjustment coefficient based on the thermal runaway mode, and determining a target thermal runaway threshold based on the target adjustment coefficient and a preset thermal runaway threshold; generating an early warning command if the temperature rise rate in a later time period is greater than the target thermal runaway threshold, the early warning command being used to instruct a battery thermal runaway early warning to be issued to the user.

[0007] According to another embodiment of the present application, an electric vehicle is also provided, including a battery and a battery management system as described above, the battery management system being used to monitor thermal runaway of the battery.

[0008] This application proposes a battery management system, including a temperature sensor and a controller, wherein the temperature sensor is communicatively connected to the controller. The temperature sensor is used to acquire temperature data of the vehicle battery in the current time period. The controller is used to: determine a thermal runaway mode based on the temperature data, the thermal runaway mode including mechanical runaway mode, electrical runaway mode, and high-temperature runaway mode; determine a target adjustment coefficient based on the thermal runaway mode; determine a target thermal runaway threshold based on the target adjustment coefficient and a preset thermal runaway threshold; and generate a warning command when the temperature rise rate in a subsequent time period exceeds the target thermal runaway threshold. The warning command is used to instruct the user to issue a battery thermal runaway warning. This embodiment uses the temperature sensor of the battery management system to collect vehicle battery temperature data in real time, identifies the thermal runaway triggering mode based on the temperature data change characteristics, dynamically determines the corresponding thermal runaway threshold for different thermal runaway modes, and generates a thermal runaway warning command when the actual temperature rise rate in a subsequent time period exceeds the corresponding thermal runaway threshold. This solves the technical problem of how to accurately provide thermal runaway warnings for vehicle batteries in related technologies, thereby improving the accuracy of vehicle battery thermal runaway judgment. Attached Figure Description

[0009] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0010] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1This is a structural block diagram of a battery management system according to an embodiment of this application;

[0012] Figure 2 This is a schematic flowchart of a battery thermal runaway early warning method according to an embodiment of this application;

[0013] Figure 3 This is a flowchart of a battery thermal runaway early warning method according to an embodiment of this application. Detailed Implementation

[0014] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0015] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0016] However, there may be instances where unnecessary detailed descriptions are omitted. For example, detailed descriptions of well-known matters or repetitive descriptions of essentially the same structure may be omitted. This is to avoid making the following description unnecessarily lengthy and to facilitate understanding by those skilled in the art. Furthermore, the following description is provided to enable those skilled in the art to fully understand this application and is not intended to limit the subject matter of the claims.

[0017] The following appropriately discloses an example of a battery according to this application. The battery in this application is a secondary battery, also known as a rechargeable battery or a storage battery, which refers to a battery that can be used again after being discharged by recharging to activate the active materials.

[0018] Typically, a secondary battery includes an electrode assembly, an electrolyte, and an outer casing. The electrode assembly consists of a positive electrode, a negative electrode, and a separator. The electrode assembly and electrolyte are assembled inside the outer casing. During charging and discharging, active ions (such as lithium ions) move back and forth between the positive and negative electrodes, inserting and extracting. The separator, positioned between the positive and negative electrodes, primarily prevents short circuits while allowing active ions to pass through. The electrolyte, located between the positive and negative electrodes, mainly serves to conduct active ions.

[0019] A Battery Management System (BMS) is an electronic system used to monitor and manage battery packs (especially large battery packs composed of multiple cells, such as those in electric vehicles and energy storage systems). Its main functions include:

[0020] Battery status monitoring: Real-time monitoring of key parameters such as battery voltage, current, temperature, and state of charge to ensure that the battery operates within a safe range.

[0021] Balanced management: By controlling the charging and discharging process of each cell in the battery pack, the energy distribution among the cells is balanced, preventing overcharging and discharging of any part, thereby extending the overall lifespan of the battery pack.

[0022] Fault diagnosis and early warning: When abnormal conditions are detected, such as overheating, overcharging, over-discharging, short circuit, etc., the BMS can react quickly, issue alarm signals and take necessary protective measures, such as suspending charging and cutting off the circuit, to prevent battery damage or safety accidents.

[0023] Safety Management: During battery pack operation, the BMS implements various safety strategies, such as overcurrent protection, thermal runaway prevention, and battery status assessment, to ensure the safe operation of the battery pack.

[0024] Communication and Control: The BMS typically communicates with the vehicle controller, the central control unit of the energy storage system, or other external devices to provide battery status information and receive control commands from the outside, such as adjusting the charging rate and reporting battery health status.

[0025] Data recording and analysis: Record the battery pack's operating data for later analysis of battery performance, prediction of lifespan, and assessment of the overall system health.

[0026] Battery Management System (BMS) is an indispensable component in applications such as electric vehicles, energy storage systems, and portable electronic devices. It plays the role of a guardian of battery pack health, ensuring that the battery operates in optimal condition through sophisticated algorithms and sensor networks, while minimizing the occurrence of safety accidents.

[0027] Figure 1This is a structural block diagram of a battery management system according to an embodiment of this application; as shown below. Figure 1 As shown, the battery management system includes:

[0028] Temperature sensor 12 and controller 14, wherein the temperature sensor 12 and the controller 14 are communicatively connected;

[0029] The temperature sensor 12 is used to acquire the temperature data of the vehicle battery during the current time period;

[0030] The controller 14 is used to determine the thermal runaway mode based on the temperature data, the thermal runaway mode including mechanical runaway mode, electrical runaway mode and high temperature runaway mode; determine the target adjustment coefficient based on the thermal runaway mode, and determine the target thermal runaway threshold based on the target adjustment coefficient and the preset thermal runaway threshold; if the temperature rise rate in a later period is greater than the target thermal runaway threshold, generate an early warning command, the early warning command is used to instruct the user to issue a battery thermal runaway warning.

[0031] Optionally, in the above embodiments, for example, the temperature sensor and controller in the battery management system establish a communication connection through the vehicle CAN (Controller Area Network) bus or vehicle Ethernet, and the temperature sensor periodically sends the collected temperature data to the controller in the form of a digital signal; or, the temperature sensor is directly connected to the controller through a hard wire connection to output the temperature value as an analog voltage signal, and the controller integrates an analog-to-digital conversion module to convert the analog quantity into a digital quantity for processing.

[0032] Temperature sensors are used to collect real-time temperature data of the vehicle battery within the current time period. For example, PT100 RTDs or K-type thermocouples are used to continuously acquire the temperature value of a specified location on the battery surface with a sampling period of 0.1 seconds. The sampling period can be customized according to actual needs. After receiving the temperature data, the controller first identifies the thermal runaway trigger mode based on the temperature data, classifying the thermal runaway mode into three types: Mechanical runaway mode refers to thermal runaway caused by external forces such as collision, squeezing, or puncture, resulting in damage to the battery structure, diaphragm breakage, or direct short circuit between the positive and negative electrodes. Its typical characteristic is a sudden spike in temperature from room temperature; Electrical runaway mode refers to thermal runaway caused by electrical abuse such as overcharging, over-discharging, external short circuits, or internal short circuits, resulting in the accumulation of internal side reactions in the battery. Its typical characteristic is a gradual increase in temperature during the charging or discharging process; High-temperature runaway mode refers to a chain reaction triggered by a continuous rise in the internal temperature of the battery due to an external high-temperature environment, heat dissipation system failure, or thermal management system failure. Its typical characteristic is that the battery temperature is initially at a high level and continues to rise.

[0033] Subsequently, the controller determines the corresponding target adjustment coefficient based on the identified thermal runaway mode. For example, adjustment coefficient K1 is used for mechanical runaway mode, adjustment coefficient K2 for electrical runaway mode, and adjustment coefficient K3 for high-temperature runaway mode. This target adjustment coefficient is then multiplied by a pre-calibrated preset thermal runaway threshold (e.g., a baseline temperature rise rate of 3℃ / s) to obtain the target thermal runaway threshold dynamically adapted to the current mode. Afterward, the controller continuously monitors the actual temperature rise rate in the next time period (the period following the current one), for example, calculating the average temperature rise rate over the past 3 seconds every second. When this temperature rise rate exceeds the target thermal runaway threshold, a warning command is immediately generated. This command is sent via the CAN bus to the instrument panel, displaying a red warning icon and triggering a buzzer. Simultaneously, it is sent to the user's mobile phone via SMS or application push through the vehicle's T-BOX (Telematics Box), and uploaded to the cloud monitoring platform. This solution significantly improves the accuracy of thermal runaway detection by distinguishing different thermal runaway modes and dynamically adjusting the detection threshold, avoiding misjudgments or missed judgments caused by a single threshold. For example, it can avoid misjudging the normal temperature rise of a vehicle heated at a large rate in a cold environment as thermal runaway, and it can also avoid missing the slow temperature rise caused by overcharging. At the same time, it achieves early warning in the early stage of thermal runaway.

[0034] The battery management system described above acquires real-time vehicle battery temperature data from temperature sensors, identifies thermal runaway triggering modes based on temperature data change characteristics, dynamically determines corresponding thermal runaway thresholds for different thermal runaway modes, and generates a thermal runaway warning command when the actual temperature rise rate in subsequent periods exceeds the corresponding thermal runaway threshold. This solves the technical problem of how to accurately provide thermal runaway warnings for vehicle batteries, thereby improving the accuracy and reliability of vehicle battery thermal runaway judgment.

[0035] In an exemplary embodiment, the controller is further configured to determine the thermal runaway mode by: determining the mechanical runaway mode as the thermal runaway mode when the temperature data indicates that the temperature value of the vehicle battery in the current time period exhibits a first temperature characteristic, wherein the first temperature characteristic indicates that the temperature value is always within a first temperature range; determining the electrical runaway mode as the thermal runaway mode when the temperature data indicates that the temperature value of the vehicle battery in the current time period exhibits a second temperature characteristic, wherein the second temperature characteristic indicates that the temperature value continues to rise and the temperature rise is greater than a preset temperature difference; and determining the high-temperature runaway mode as the thermal runaway mode when the temperature data indicates that the temperature value of the vehicle battery in the current time period exhibits a third temperature characteristic, wherein the third temperature characteristic indicates that the temperature value is always within a second temperature range, wherein the lower limit of the second temperature range is greater than the upper limit of the first temperature range.

[0036] Optionally, in the above embodiments, the controller determines the thermal runaway mode based on the temperature characteristics presented by the temperature data. Specifically, if the temperature value within the current time period (e.g., a 30-second observation window) remains within a first temperature range, which is a normal temperature range, such as 20°C to 40°C, it is determined to be a mechanical runaway mode. This is because mechanical abuse causes instantaneous damage to the internal structure of the battery, but the heat has not yet been fully conducted to the sensor location, and the temperature has not yet risen significantly. For example, when the battery is squeezed after a vehicle collision but has not yet short-circuited and heated up, the temperature sensor still measures a normal temperature value. If the temperature value shows a continuous upward trend, that is, the temperature monotonically increases within a continuous time window, and the increase exceeds a preset temperature difference, the preset temperature difference can be set to 5°C to 10°C depending on the battery type. For example, the temperature rises from 30°C to 40°C within 30 seconds. If the temperature rises to 41°C, then increases by 11°C, exceeding the 10°C threshold, it is determined to be an electrical runaway mode. This corresponds to the typical characteristics of the battery's internal side reactions gradually intensifying and heat continuously accumulating during overcharging. For example, when the battery is overcharged, the electrolyte decomposes and generates heat, causing the temperature to rise steadily. If the temperature value remains in the second temperature range, which is a high-temperature range, such as above 50°C, and the lower limit of this range (50°C) is greater than the upper limit of the first temperature range (40°C), it is determined to be a high-temperature runaway mode. This corresponds to scenarios where the battery is in a high-temperature environment for a long time or where heat dissipation fails. For example, after a vehicle has been exposed to high temperatures in summer, the battery temperature has reached 65°C. If the cooling fan fails at this time, the battery will continue to heat up on top of the high temperature.

[0037] Through the rapid pattern recognition based on temperature characteristics, the causes of thermal runaway can be preliminarily determined without complex calculations, laying the foundation for subsequent differentiated threshold setting and improving the speed of early warning response.

[0038] In an exemplary embodiment, the controller is further configured to determine the thermal runaway mode by: when it is determined that the temperature data indicates that the temperature value of the vehicle battery in the current time period exhibits a fourth temperature characteristic, inputting the temperature data into a thermal runaway mode prediction model to obtain a thermal runaway mode prediction result output by the thermal runaway mode prediction model, wherein the fourth temperature characteristic is a temperature characteristic other than the first temperature characteristic, the second temperature characteristic, and the third temperature characteristic, and the thermal runaway mode prediction model is a pre-trained model used to predict the probability that the output temperature data conforms to the temperature characteristics of each thermal runaway mode; and determining the thermal runaway mode based on the thermal runaway mode with the highest probability in the thermal runaway mode prediction result.

[0039] Optionally, in the above embodiments, when the temperature characteristics presented by the temperature data cannot be clearly classified into the first, second, or third temperature characteristics, a fourth temperature characteristic appears. For example, the temperature may fluctuate irregularly, rise first, then fall, then rise again, or rise in a stepwise manner. The controller inputs the temperature data for the current time period into a pre-trained thermal runaway mode prediction model. This model is trained based on a large amount of historical experimental data, covering temperature-time series under different battery types, different SOCs, different aging levels, and different abuse conditions. The model structure can employ a Long Short-Term Memory (LSTM) network or a Convolutional Neural Network (CNN). The input layer receives the temperature sequence within a continuous time window (e.g., one sampling point every 0.1 seconds over the past 60 seconds, for a total of 600 points). The output layer consists of three neurons, outputting the probability values ​​of the temperature data conforming to the temperature characteristics of mechanical, electrical, and high-temperature thermal runaway modes, respectively. The sum of these probability values ​​is 1. The controller selects the thermal runaway mode with the highest probability as the final identification result.

[0040] For example, if the battery temperature rises slightly after being subjected to minor pressure but then stabilizes, and then begins to rise slowly again after 30 seconds, the model might output a 60% probability of mechanical mode, a 30% probability of electrical mode, and a 10% probability of high-temperature mode, thus still classifying it as a mechanical runaway mode. If the battery experiences an internal micro-short circuit under high-temperature conditions, resulting in a fluctuating temperature rise on top of a high-temperature base, the model might output a 55% probability of high-temperature mode, a 40% probability of mechanical mode, and a 5% probability of electrical mode, thus classifying it as a high-temperature runaway mode. This solution utilizes machine learning algorithms to handle complex and ambiguous temperature changes, such as identifying complex scenarios like delayed internal short circuits caused by mechanical damage, significantly enhancing the robustness and adaptability of thermal runaway mode recognition, ensuring accurate identification of the thermal runaway type under various complex operating conditions.

[0041] In an exemplary embodiment, the controller is further configured to determine the target adjustment coefficient by: acquiring the current state of charge (SOC) value of the vehicle battery; determining a target mapping relationship from a plurality of mapping relationships based on the current SOC value, wherein the plurality of mapping relationships are used to record adjustment coefficients preset for different thermal runaway modes of the vehicle battery under different SOC values, and the target mapping relationship is used to record adjustment coefficients preset for the vehicle battery under the current SOC value; and determining the target adjustment coefficient based on the target mapping relationship.

[0042] Optionally, in the above embodiments, after determining the thermal runaway mode, the controller also needs to combine the current state of charge (SOC) of the battery to obtain the target adjustment coefficient. The system internally stores multiple mapping tables in the controller's non-volatile memory. Each table corresponds to a specific SOC interval, for example, dividing the SOC from 0% to 100% into five intervals: 0%–20%, 20%–40%, 40%–60%, 60%–80%, and 80%–100%, with each interval corresponding to an independent mapping table; or using a finer division, with each interval representing 5% or 10%. Each table records the preset adjustment coefficients for different thermal runaway modes within that SOC interval. These coefficients are obtained through statistical fitting of a large amount of thermal runaway experimental data.

[0043] The controller first obtains the current SOC value through the BMS's SOC estimation module. For example, if the current SOC is 75% using the ampere-hour integral method combined with open-circuit voltage correction, then the target mapping table corresponding to 75% SOC is selected from multiple mapping tables, i.e., the table corresponding to the 60%–80% range. Then, based on the determined thermal runaway mode (such as electrical runaway mode), the corresponding adjustment coefficient is found from the target mapping table. These mapping relationships reflect the differences in battery thermal runaway sensitivity under different SOCs. For example, at high SOCs (above 80%), the battery has high lithium intercalation in the negative electrode, high lithium activity, and is more prone to thermal runaway, with a lower critical temperature rise rate. Therefore, the adjustment coefficient is usually taken as a small value (e.g., 0.7–0.9) to lower the threshold and achieve early warning. At low SOCs (below 20%), the battery has low energy density, relatively good thermal stability, and a higher critical temperature rise rate. The adjustment coefficient is taken as a larger value (e.g., 1.1–1.3) to avoid false alarms at low charge levels. By dynamically adjusting the SOC, the threshold is more closely aligned with the real-time battery status, further improving the accuracy of the judgment. For example, it can avoid missed judgments due to excessively high thresholds when the battery is fully charged, and also avoid false alarms due to excessively low thresholds when the battery is depleted.

[0044] In an exemplary embodiment, the controller is further configured to determine the target thermal runaway threshold by: obtaining a first adjustment coefficient corresponding to the mechanical runaway mode, a second adjustment coefficient corresponding to the electrical runaway mode, and a third adjustment coefficient corresponding to the high-temperature runaway mode from the target mapping relationship; if the thermal runaway mode is determined to be the mechanical runaway mode, determining the first adjustment coefficient as the target adjustment coefficient; or, if the thermal runaway mode is determined to be the electrical runaway mode, determining the second adjustment coefficient as the target adjustment coefficient; or, if the thermal runaway mode is determined to be the high-temperature runaway mode, determining the third adjustment coefficient as the target adjustment coefficient; and determining the target thermal runaway threshold by multiplying the target adjustment coefficient by the preset thermal runaway threshold.

[0045] Optionally, in the above embodiments, the controller obtains the adjustment coefficients corresponding to the three thermal runaway modes—mechanical, electrical, and high-temperature—from the target mapping table, respectively, and denoted as the first adjustment coefficient k1, the second adjustment coefficient k2, and the third adjustment coefficient k3. Taking the mapping table corresponding to the SOC 60%–80% range as an example, assume that the table records k1=1.15, k2=1.05, and k3=0.95. When the thermal runaway mode determined in the steps is the mechanical runaway mode, the controller assigns k1 as the target adjustment coefficient through logical judgment; when it is the electrical runaway mode, k2 is used as the target adjustment coefficient; and when it is the high-temperature runaway mode, k3 is used as the target adjustment coefficient. Then, the controller calculates the product of the target adjustment coefficient and the preset thermal runaway threshold T_base to obtain the final target thermal runaway threshold used for judgment. T_base is a benchmark value calibrated according to the battery model and chemical system, for example, 3℃ / s for ternary lithium batteries and 2℃ / s for lithium iron phosphate batteries. Assuming T_base = 3℃ / s, if the current runaway mode is electrical and k2 = 1.05, the target threshold is 3.15℃ / s; if the current runaway mode is high temperature and k3 = 0.95, the target threshold is 2.85℃ / s; and if the current runaway mode is mechanical and k1 = 1.15, the target threshold is 3.45℃ / s. This calculation process directly links the threshold with the thermal runaway mode and SOC, ensuring the real-time adaptability of the judgment criteria.

[0046] In an exemplary embodiment, the controller is further configured to determine the target adjustment coefficient by: determining the thermal runaway mode as a multidimensional thermal runaway mode when the thermal runaway mode prediction result indicates that the probability of at least two thermal runaway modes is greater than a preset probability; and determining the target adjustment coefficient based on the probability of each thermal runaway mode among the at least two thermal runaway modes and the adjustment coefficient corresponding to each thermal runaway mode.

[0047] Optionally, in the above embodiments, when the output of the thermal runaway mode prediction model shows that the probabilities of two or more modes exceed a preset probability threshold (e.g., 40%), the controller determines that the current thermal runaway is a multidimensional thermal runaway mode, meaning that the battery may be simultaneously affected by the superposition of multiple abuse factors, such as an internal short circuit occurring under high temperature conditions (mechanical and high temperature combination), or slight deformation of the battery during overcharging (electrical and mechanical combination). In this case, a single-mode adjustment coefficient cannot be simply used, because a single coefficient cannot accurately reflect the actual thermal runaway critical conditions under combined effects. The controller first identifies all modes whose probabilities exceed the threshold, such as a mechanical mode probability of 45% and a high temperature mode probability of 40%, both exceeding the 40% threshold; then it obtains the adjustment coefficients corresponding to these modes (k1=1.15, k3=0.95); and then, combined with the probability values ​​of each mode, it jointly determines a comprehensive target adjustment coefficient. This solution can effectively handle common complex thermal runaway scenarios in practical applications. For example, if a vehicle collides with another vehicle in a high-temperature environment in summer and the battery catches fire with a delay, it is a combination of high temperature and mechanical factors. The system can avoid threshold deviations caused by simply classifying it as mechanical or high temperature through multi-dimensional pattern recognition, thereby improving the system's adaptability and reliability.

[0048] In an exemplary embodiment, the controller is further configured to determine the target adjustment coefficient by using the probabilities of each thermal runaway mode as weighting coefficients and summing the adjustment coefficients corresponding to each thermal runaway mode to obtain the target adjustment coefficient.

[0049] Optionally, in the above embodiments, for multidimensional thermal runaway modes, the controller calculates the target adjustment coefficient using a weighted summation method: the probability of each mode exceeding the probability threshold is used as a weight, multiplied by its corresponding adjustment coefficient, and then the products are summed to obtain the final target adjustment coefficient. Taking a mechanical mode probability P1=0.45 and adjustment coefficient k1=1.15, and a high-temperature mode probability P3=0.40 and adjustment coefficient k3=0.95 as an example, the target adjustment coefficient = (0.45×1.15+0.40×0.95) / (0.45+0.40)=1.056. Dividing by the probability sum is for normalization, ensuring that the comprehensive coefficient reflects the weighted average of each mode. This probability-based weighted fusion method scientifically reflects the comprehensive influence of different factors on battery thermal runaway behavior, enabling the target threshold to more accurately match actual complex operating conditions. For example, when mechanical and thermal factors each account for half, the comprehensive threshold lies between the two, significantly improving the accuracy of the judgment.

[0050] In an exemplary embodiment, the battery management system further includes a data acquisition module, which is configured to: acquire experimental data of thermal runaway experiments conducted on the vehicle battery under different states of charge, wherein the thermal runaway experiments are used to test the thermal runaway data of the vehicle battery under different thermal runaway models; and determine the preset thermal runaway threshold and the plurality of mapping relationships based on the experimental data.

[0051] Optionally, in the above embodiments, the battery management system further includes a data acquisition module, which can be used to acquire experimental data generated from thermal runaway experiments of the vehicle battery under different states of charge. Specifically, for the same type of battery, thermal runaway is triggered under different SOC conditions (e.g., 0%, 25%, 50%, 75%, 100%) through mechanical abuse, electrical abuse, and thermal abuse, and parameters such as the critical temperature rise rate, critical temperature, and voltage drop moment at the time of thermal runaway are recorded. Experimental data includes temperature-time curves, voltage-time curves, pressure changes, etc., and at least three repeated experiments are performed for each SOC point and each mode to ensure statistical validity. Based on these experimental data, statistical analysis or machine learning regression methods are used to determine the preset thermal runaway threshold (e.g., the median or mean of the critical temperature rise rate of all experimental data) and the mapping relationship of the adjustment coefficient under different SOCs and different thermal runaway modes, for example, by fitting the functional relationship between SOC and adjustment coefficient. These data are the basis for the offline calibration of the system, ensuring the scientific validity and accuracy of the thresholds and coefficients, and providing a reliable basis for subsequent online applications.

[0052] In an exemplary embodiment, the data acquisition module is further configured to acquire the experimental data in the following ways: acquire first experimental data corresponding to the mechanical runaway mode, wherein the first experimental data is thermal runaway data of the vehicle battery under mechanical damage to the battery structure at different states of charge; acquire second experimental data corresponding to the electrical runaway mode, wherein the second experimental data is thermal runaway data of the vehicle battery under abnormal electrical parameters at different states of charge; acquire third experimental data corresponding to the high-temperature runaway mode, wherein the third experimental data is thermal runaway data of the vehicle battery under different states of charge in a third temperature range; and determine the experimental data based on the first experimental data, the second experimental data, and the third experimental data.

[0053] Optionally, in the above embodiments, when acquiring experimental data, the data acquisition module conducts specific experiments for three thermal runaway modes: For the mechanical runaway mode, mechanical damage is simulated by applying pressure (pressure gradually increasing from 0kN to 200kN), needle puncture (steel needle diameter 3mm, 5mm, 8mm, puncture speed 0.1mm / s to 10mm / s), or heavy object impact (mass 10kg, drop height 1m) under different SOC states of the battery, and temperature, voltage, and pressure data are collected during the thermal runaway process, and the critical temperature rise rate at the trigger of thermal runaway is recorded; For the electrical runaway mode, electrical anomalies are simulated by overcharging, over-discharging, or external short circuits, and temperature rise changes and critical conditions are recorded; For the high-temperature runaway mode, the battery is placed in a high-temperature environment chamber, and the ambient temperature is set to rise from 50℃ to 200℃ at different rates (1℃ / min, 5℃ / min, 10℃ / min) to simulate heat dissipation failure or external fire, and the temperature rise rate and temperature at the trigger of thermal runaway are monitored. The experimental data obtained from the first, second, and third modes are combined to form a complete experimental dataset. Each data point includes information such as SOC value, thermal runaway mode label, critical temperature rise rate, and critical temperature. This experimental design comprehensively covers various abuse scenarios that may occur in real-world applications. For example, the needle penetration test simulates puncture of the vehicle's undercarriage, the overcharge test simulates charging pile failure, and the hot box test simulates a vehicle fire, ensuring the completeness and representativeness of the mapping table.

[0054] In an exemplary embodiment, the controller is further configured to generate the warning instruction by: calculating the temperature rise rate using a time sliding window in the subsequent time period, wherein the time window size of the time sliding window is a preset duration, and the temperature rise rate is the rate of temperature increase within the time window; and generating the warning instruction if it is determined that the temperature rise rate is always greater than the target thermal runaway threshold during the preset sliding duration of the time sliding window.

[0055] Optionally, in the above embodiments, the controller uses a time-sliding window technique to calculate the temperature rise rate to avoid instantaneous noise interference. Specifically, a time window is set, the size of which can be configured to 3 seconds, 5 seconds, or 10 seconds. The temperature data continuously collected within the window is fitted using linear least squares to obtain the temperature rise rate, i.e., the slope of the fitted line, or the temperature difference between the beginning and end of the window is calculated by dividing the window duration by the simple difference method. For example, with a 5-second window, 10 points are collected per second (sampling period of 0.1 seconds), resulting in 50 temperature points within the window. The average temperature rise rate of the window is then fitted. The window is then moved with a preset sliding step size (e.g., 1 second), i.e., every second, the window slides forward by 1 second, discarding the oldest data points and adding the newest data points, and the temperature rise rate of the new window is recalculated. Thermal runaway is determined to have occurred and an early warning command is generated only when the temperature rise rates of multiple consecutive windows (e.g., 3 consecutive windows) are all greater than the target thermal runaway threshold. For example, if the target threshold is 3℃ / s, and the calculated values ​​for three consecutive 5-second windows are 3.2℃ / s, 3.5℃ / s, and 3.3℃ / s respectively, all exceeding the threshold, then an alert is triggered. If only a single window exceeds the threshold, it may be noise or a brief fluctuation, and no alert is triggered. This dual judgment mechanism of sliding window and continuous threshold exceedance effectively filters out sensor spike interference (such as instantaneous jumps caused by electromagnetic interference) or brief temperature fluctuations (such as temperature fluctuations caused by coolant pulsation), significantly reducing the false alarm rate and improving the reliability of the alert.

[0056] In one exemplary embodiment, the temperature sensor is positioned at the geometric center of the largest surface of the vehicle battery not covered by the heating element, and is used to collect the temperature data at a preset sampling period.

[0057] Optionally, in the above embodiments, the temperature sensor is positioned at the geometric center of the largest surface of the vehicle battery not covered by the heating element (i.e., the large surface of the battery). For square aluminum-cased batteries, the largest surface typically consists of two sides with the largest area, and the sensor is attached to the center point of one of these sides; for pouch batteries, it is similarly attached to the center of the large surface. This position most accurately reflects the average temperature change inside the battery and is unaffected by thermal interference from local heating elements. The sensor can be an epoxy-encapsulated thermistor or a thin-film thermocouple, tightly bonded to the battery surface with thermally conductive adhesive to ensure rapid thermal response. The sensor continuously collects temperature data at a preset sampling period, which can be set to 0.1 seconds, i.e., 10 temperature points are collected per second. This frequency is sufficient to capture the details of the dramatic temperature changes during thermal runaway (the temperature rise rate during thermal runaway can reach over 100°C / s, and 10 points can be collected at a 0.1-second interval, which is sufficient to reconstruct the change curve). If a higher sampling frequency, such as 0.01 seconds, is used, even finer transient changes can be captured. This sampling method and frequency enable the system to capture the most subtle temperature changes in the early stages of thermal runaway, providing a high-quality data foundation for subsequent temperature rise rate calculation and pattern recognition, thereby improving the overall monitoring accuracy.

[0058] In one exemplary embodiment, the battery management system further includes a flame sensor and a smoke sensor, both of which are communicatively connected to the controller. After generating the warning command, the controller is further configured to: disconnect the high-voltage circuit of the vehicle battery to stop charging or heating the vehicle battery; control the vehicle battery to actively discharge at a constant power through a preset discharge resistor until the voltage of the vehicle battery drops below a safe voltage threshold; and, if the flame sensor detects a fire in the installation space of the vehicle battery, and / or if the smoke sensor detects smoke in the installation space of the vehicle battery, send a fire extinguishing command to the battery thermal management system to activate the fire extinguishing device.

[0059] Optionally, in the above embodiments, after the controller generates a warning command, it immediately executes multi-level interlocking protection actions: First, by controlling the high-voltage contactor to disconnect, the high-voltage circuit of the vehicle battery is cut off, making the battery electrically isolated from external equipment such as the drive motor and charging pile. At the same time, the internal control logic of the BMS stops any ongoing charging or heating operations, such as turning off the on-board charger or disconnecting the power supply to the heating film, blocking energy input from the source. Second, the battery is controlled to actively discharge through a preset discharge resistor. The discharge resistor can be a power resistor (such as 100Ω, 500W), connected between the positive and negative terminals of the battery. The BMS controls a discharge relay to close, so that the battery discharges at a constant power (such as 500W). During the discharge process, the BMS continuously monitors the battery voltage. When the voltage drops below the safe voltage threshold (such as below 60V, which meets the human body safety voltage standard), the discharge relay is disconnected, thereby reducing the risk of subsequent short circuit or fire. At the same time, the system also integrates a flame sensor and a smoke sensor. The flame sensor can be an infrared flame detector with a response time of less than 1 second, and the smoke sensor can be a photoelectric smoke detector to detect particulate matter concentration. When the flame sensor detects flame radiation of a specific wavelength within the installation space, or the smoke sensor detects smoke concentration exceeding a preset threshold, the controller sends a fire extinguishing command to the battery thermal management system (TMS). The TMS then activates the fire extinguishing devices, such as opening the release valve of an aerosol fire extinguisher, precisely spraying the extinguishing agent into the thermal runaway cell area through pipes installed within the battery pack. This series of actions forms a complete protection chain from electrical isolation and energy release to physical fire extinguishing, minimizing the spread of thermal runaway hazards. For example, cutting off high voltage in the early stages of thermal runaway prevents external short circuits from causing secondary accidents, active discharge reduces residual battery energy, and the fire extinguishing device extinguishes open flames to prevent heat spread, protecting the vehicle and its occupants.

[0060] In an exemplary embodiment, after generating the warning command, the controller is further configured to: activate a high-speed data acquisition mode to record multidimensional data of the vehicle battery during thermal runaway at a preset sampling frequency; align the multidimensional data according to the time axis to generate a data packet containing a timestamp, a data identifier, and a data checksum; and upload the data packet to a cloud server and simultaneously save it to a local storage device.

[0061] Optionally, in the above embodiments, after generating the warning command, the controller immediately activates the high-speed data acquisition mode, switching the conventional sampling frequency (e.g., 1Hz) to a preset high sampling frequency, such as 100Hz or higher, to simultaneously record multi-dimensional data of the battery before and after thermal runaway. The multi-dimensional data includes: temperature at each monitoring point, voltage of each cell, total current, temperature rise rate, SOC, insulation resistance, etc. All data are collected using the same time reference to ensure time axis alignment. Subsequently, the system aligns these multi-dimensional data according to the time axis and generates a data packet. The data packet format adopts a standardized binary or JSON format. Each data point contains a precise timestamp (e.g., millisecond-level Unix timestamp), data identifier, data value, and data checksum. The checksum is used to ensure the integrity of data transmission and storage. After generating the data packet, the controller uploads the data packet to the cloud server via the vehicle-to-grid (T-BOT); simultaneously, it saves the data packet to local storage, such as writing it to a read-only partition of a dedicated black box storage device, and generates a write log, recording information such as the save time, file size, and checksum results, for offline retrieval after an accident. This data recording mechanism provides valuable original evidence for subsequent accident tracing (such as reconstructing the thermal runaway process), model optimization (such as correcting the mapping table), and insurance claims, which helps to continuously improve battery safety technology.

[0062] In an optional embodiment, Figure 2 This is a schematic flowchart of a battery thermal runaway early warning method according to an embodiment of this application, as shown below. Figure 2 As shown, the specific steps include:

[0063] Step S201: Temperature Data Acquisition; After system startup, the temperature sensor continuously acquires temperature data of the battery surface according to a preset sampling period (0.1 seconds in this embodiment). The sensor is positioned at the geometric center of the largest surface of the battery not covered by the heating element. This location most accurately reflects the average temperature change inside the battery and is unaffected by localized heating or cooling. The acquired temperature data is transmitted to the controller in real time via a CAN bus or hardwired connection, serving as the basis for all subsequent analyses.

[0064] Step S202: Thermal runaway mode identification; After receiving temperature data for the current time period (e.g., the past 30 seconds), the controller first identifies the thermal runaway trigger mode based on temperature change characteristics. The identification process involves three typical scenarios:

[0065] If the temperature value remains within the first temperature range (normal temperature range, such as 20℃~40℃), it indicates that the battery has not been affected by heat. It may be a sudden thermal runaway caused by mechanical damage, which is determined to be a mechanical runaway mode.

[0066] If the temperature value shows a continuous upward trend and the increase exceeds the preset difference (e.g., the temperature rise exceeds 10°C within 30 seconds), it indicates that cumulative heat generation is occurring inside the battery, which is determined to be an electrical runaway mode.

[0067] If the temperature value remains in the second temperature range (high temperature range, such as above 50°C), it indicates that the battery is in a high temperature environment and is determined to be in high temperature runaway mode.

[0068] Step S203: Composite Mode Determination; For cases with atypical temperature characteristics, the system inputs the temperature sequence into the thermal runaway mode prediction model to obtain the probability output of each mode. When the probability of two or more modes exceeds a preset threshold (e.g., 40%), it is determined to be a multidimensional thermal runaway mode, meaning the battery is simultaneously subjected to the combined effects of multiple factors. In this case, the system uses a weighted summation method to calculate the comprehensive regulation coefficient, using the probability of each mode as a weight, and performs a weighted average of its corresponding regulation coefficients to obtain the comprehensive regulation coefficient adapted to the composite scenario.

[0069] Step S204: SOC Acquisition; The system acquires the current state of charge (SOC) value of the battery through the SOC estimation module of the BMS. SOC estimation can employ algorithms such as the ampere-hour integration method, open-circuit voltage correction method, or Kalman filtering to ensure real-time performance and accuracy. The SOC value reflects the current energy state of the battery and is a crucial factor affecting the critical conditions for thermal runaway.

[0070] Step S205: Mapping Relationship Table Lookup; The system pre-stores multiple mapping relationship tables, each corresponding to a SOC range (e.g., 0-20%, 20-40%, 40-60%, 60-80%, 80-100%). Each table records the preset adjustment coefficients for different thermal runaway modes within that SOC range. These coefficients are statistically calibrated based on a large amount of thermal runaway experimental data, reflecting the differences in battery thermal runaway sensitivity under different SOCs. The system selects the corresponding target mapping relationship table based on the current SOC value and, in conjunction with the determined thermal runaway mode, reads the corresponding adjustment coefficients from it.

[0071] Step S206: Calculation of the target thermal runaway threshold; The system multiplies the adjustment coefficient obtained in step S205 with the preset thermal runaway threshold (e.g., a reference temperature rise rate of 3℃ / s) to obtain the target thermal runaway threshold that dynamically adapts to the current battery state and thermal runaway mode. This threshold will be used as the basis for judgment in subsequent real-time monitoring.

[0072] Step S207: Calculation of temperature rise rate in a sliding window; the system uses a time sliding window technique to calculate the temperature rise rate for the next time period to avoid instantaneous noise interference. Specifically, a time window (e.g., 5 seconds) is set, and the average temperature rise rate is obtained by linear fitting of the continuously collected temperature data within the window; then, the window is moved with a preset sliding step size (e.g., 1 second), and the temperature rise rate of the new window is continuously calculated. This method can smooth temperature fluctuations and improve the stability of the calculation results.

[0073] Step S208: Threshold Exceedance Determination; The system compares the temperature rise rate calculated for each sliding window with the target thermal runaway threshold determined in step S206. When the temperature rise rate for multiple consecutive windows (e.g., three consecutive windows) exceeds the target threshold, thermal runaway is determined to be imminent or has already occurred, and the process proceeds to step S209; otherwise, it returns to step S207 to continue monitoring. This continuous threshold exceeding determination condition effectively filters out transient interference and improves the reliability of the early warning.

[0074] Step S209: Generate a warning command; when the judgment conditions are met, the system immediately generates a warning command. This command is sent to the instrument panel via the CAN bus, triggering visual and audible alarms; simultaneously, warning information is pushed to the user's mobile phone via the in-vehicle T-BOX, and alarm data is uploaded to the cloud monitoring platform to achieve multi-channel warning.

[0075] Step S210: Execute multi-level interlocking protection; after the warning command is generated, the system synchronously starts multi-level interlocking protection actions:

[0076] Level 1: Disconnect the high-voltage circuit and disconnect the main positive / main negative contactor to electrically isolate the battery from external equipment;

[0077] Level 2: Stop all charging and heating operations, and block energy input;

[0078] Level 3: Start the active discharge circuit and discharge at a constant power through the discharge resistor until the voltage drops below the safety threshold.

[0079] Level 4: When the smoke sensor or flame sensor detects a fire or smoke, it sends a fire extinguishing command to the battery thermal management system and activates the fire extinguishing device for precise spraying.

[0080] Step S211: Initiate high-speed data acquisition and recording; while generating the warning command, the system switches the sampling frequency from the normal mode to the high-speed mode (e.g., from 1Hz to 100Hz), simultaneously recording multi-dimensional data of the entire thermal runaway process, including temperature, voltage, current, temperature rise rate, and protection action timestamps. The system aligns this data along the timeline, generates a data packet containing timestamps, data identifiers, and checksums, and uploads it to the cloud server via the vehicle communication module, while simultaneously writing it to the local black box memory to ensure data integrity and traceability.

[0081] Step S212: End; After the system completes the warning, protection, and data recording, the process ends. If the battery state stabilizes, the system can be reset and return to the initial monitoring state; if thermal runaway continues to develop, the system remains in a protected state until energy is depleted or external intervention occurs.

[0082] In one optional embodiment, the complete workflow of the battery management system of this application is illustrated by an example. For instance, an electric vehicle is equipped with a battery pack consisting of ternary lithium-ion cells with a rated capacity of 100Ah and a rated voltage of 350V. After driving at high speed for one hour in a high-temperature summer environment, the vehicle is parked in an open-air parking lot. At this time, the ambient temperature reaches 38°C, and the battery is in a high-temperature state due to continuous high-rate discharge and external high temperature. The battery management system monitors the temperature of each cell in real time through temperature sensors. The sensors are arranged as described above at the geometric center of the large surface of each cell that is not covered by the heating element, and continuously collect temperature data at a sampling period of 0.1 seconds. When the vehicle is parked, the system records an average battery pack temperature of 50°C.

[0083] While the vehicle was parked, a minor collision occurred when an adjacent vehicle reversed, causing the side of the parked vehicle to be squeezed and the battery pack casing to deform slightly but not rupture. At the moment of impact, some cell separators inside the battery suffered minor damage, but no direct short circuit occurred. Temperature sensor data showed that within 30 seconds of the collision, the temperature of each cell remained between 50°C and 52°C, without any sudden spike. System analysis of the temperature data for that period revealed that the temperature remained consistently within the high-temperature range of around 50°C, with the lower limit of 50°C exceeding the upper limit of the normal temperature range of 40°C. Based on temperature characteristic recognition logic, the system initially identified the thermal runaway mode as a high-temperature runaway mode.

[0084] To further confirm this, the system input the temperature sequence 60 seconds before and after the collision into a pre-trained thermal runaway mode prediction model. This model, trained on a large amount of historical experimental data, is capable of identifying subtle pattern features in temperature changes. The model output showed a 55% probability of a high-temperature runaway mode, a 40% probability of a mechanical runaway mode, and a 5% probability of an electrical runaway mode. Since the probability of two of these modes both exceeded 30%, the system determined the current situation to be a multidimensional thermal runaway mode, i.e., a combination of high-temperature background and mechanical damage.

[0085] Subsequently, the system obtains the current battery state of charge (SOC) value as 82%. The system retrieves the target mapping table corresponding to the SOC range of 80% to 100% from memory. This table, pre-calibrated based on extensive thermal runaway experimental data, records the adjustment coefficients corresponding to different thermal runaway modes within this SOC range: mechanical runaway mode adjustment coefficient 0.85, electrical runaway mode adjustment coefficient 0.95, and high-temperature runaway mode adjustment coefficient 0.90. Since the current thermal runaway mode is multidimensional, with a mechanical mode probability of 40% and a high-temperature mode probability of 55%, the system uses a weighted summation method to calculate the target adjustment coefficient: the probability of each mode is used as a weight, multiplied by its corresponding adjustment coefficient, and then divided by the sum of probabilities for normalization, resulting in a target adjustment coefficient of 0.879. The preset thermal runaway threshold, calibrated according to the battery model, is 3.0℃ / s. The system multiplies the target adjustment coefficient by the preset threshold to obtain a target thermal runaway threshold of 2.637℃ / s for dynamically adapting to the current mode.

[0086] After the vehicle was towed to the repair shop, the technicians connected it to a charging station for charging and testing. During charging, the micro-damage to the diaphragm caused by the collision gradually worsened under the influence of the current, leading to localized micro-short circuits and increased internal heat generation in the battery. The system used a time-sliding window technique to continuously calculate the temperature rise rate: the time window size was set to 5 seconds, the sliding step size was 1 second, and the average temperature rise rate was obtained by linear fitting of 50 temperature points within each window. After 120 seconds of charging, the system calculated the temperature rise rates for three consecutive sliding windows to be 2.5℃ / s, 2.7℃ / s, and 2.8℃ / s, respectively, all exceeding the target thermal runaway threshold of 2.637℃ / s. After meeting the condition of exceeding the threshold for multiple consecutive windows, the system immediately generated a warning command, sent it to the instrument panel via the CAN bus to display a red warning icon and trigger a voice prompt, pushed the warning information to the user's mobile phone via the in-vehicle T-BOX, and uploaded the alarm data to the cloud monitoring platform.

[0087] Upon generation of the warning command, the system immediately executes multi-level interlocking protection actions: First, it sends a disconnect command to the high-voltage contactor, cutting off the connection between the battery pack and external equipment such as the charging pile and drive motor within 20 milliseconds, while simultaneously stopping all charging and heating operations; Second, it closes the active discharge circuit, starting to discharge at a constant power through a preset discharge resistor, continuously monitoring the battery voltage until it drops below the 60V safety threshold; Third, as the internal short circuit intensifies, the battery temperature continues to rise, the cell explosion-proof valve opens, and a large amount of smoke is generated. At this time, the smoke sensor installed in the battery pack detects that the smoke concentration reaches 8% / m³, exceeding the preset threshold, and the flame sensor detects the flame characteristics generated by the cell eruption. Upon receiving the sensor signal, the system immediately sends a fire extinguishing command to the battery thermal management system; Fourth, the battery thermal management system activates the fire extinguishing device, precisely spraying the extinguishing agent to the thermal runaway cell area through pipes arranged in the battery pack, while simultaneously starting the coolant pump to run at maximum speed and activating the cooling mode to enhance cooling of the unrunaway cells and prevent heat spread.

[0088] While generating the warning command, the system activates a high-speed data acquisition mode, switching the sampling frequency from the conventional 1Hz to 100Hz. It simultaneously records multi-dimensional data of the entire thermal runaway process, including time-series temperature data at each monitoring point, time-series voltage data for each cell, total current, insulation resistance, real-time temperature rise rate, thermal runaway mode identification results and probabilities, dynamic change trajectory of the target thermal runaway threshold, and timestamps of various protection actions. The system aligns this multi-dimensional data along a unified timeline, generating a data packet containing timestamps, data identifiers, and data checksums. This packet is then prioritized for uploading to the cloud server via the vehicle-mounted T-BOX and simultaneously written to the read-only partition of the local black box memory.

[0089] In the above embodiments, through thermal runaway mode recognition and multi-dimensional weighted fusion, the battery management system accurately calculates the dynamic thermal runaway threshold adapted to the combined high temperature and mechanical conditions, issues timely and accurate warnings, and the multi-level interlocking protection completes high voltage cutoff, active discharge and fire extinguishing device activation in a short time, controlling the thermal runaway range; at the same time, the 100Hz high-frequency data record completely restores the accident process, providing accurate basis for subsequent traceability analysis, and significantly improving the safety, reliability and traceability of battery use.

[0090] This embodiment provides a battery thermal runaway early warning method, such as... Figure 3 As shown, Figure 3 This is a flowchart illustrating a battery thermal runaway early warning method according to an embodiment of this application, including the following steps:

[0091] Step S302: Obtain the temperature data of the vehicle battery during the current time period;

[0092] Step S304: Determine the thermal runaway mode based on the temperature data. The thermal runaway mode includes mechanical runaway mode, electrical runaway mode, and high temperature runaway mode.

[0093] Step S306: Determine the target adjustment coefficient according to the thermal runaway mode, and determine the target thermal runaway threshold according to the target adjustment coefficient and the preset thermal runaway threshold;

[0094] Step S308: If the temperature rise rate in the subsequent period is greater than the target thermal runaway threshold, a warning instruction is generated. The warning instruction is used to instruct the user to issue a battery thermal runaway warning.

[0095] Through the above steps, real-time vehicle battery temperature data collected by temperature sensors is obtained. Based on the characteristics of temperature data changes, the thermal runaway triggering mode is identified, and the corresponding thermal runaway threshold is dynamically determined for different thermal runaway modes. When the actual temperature rise rate in subsequent periods exceeds the corresponding thermal runaway threshold, a thermal runaway warning command is generated. This solves the technical problem of how to accurately provide thermal runaway warnings for vehicle batteries in related technologies, thereby improving the accuracy and reliability of vehicle battery thermal runaway judgment.

[0096] In one exemplary embodiment, determining a thermal runaway mode based on the temperature data includes: determining the mechanical runaway mode as the thermal runaway mode when the temperature data indicates that the temperature value of the vehicle battery in the current time period exhibits a first temperature characteristic, wherein the first temperature characteristic indicates that the temperature value is always within a first temperature range; determining the electrical runaway mode as the thermal runaway mode when the temperature data indicates that the temperature value of the vehicle battery in the current time period exhibits a second temperature characteristic, wherein the second temperature characteristic indicates that the temperature value continues to rise and the temperature rise is greater than a preset temperature difference; and determining the high-temperature runaway mode as the thermal runaway mode when the temperature data indicates that the temperature value of the vehicle battery in the current time period exhibits a third temperature characteristic, wherein the third temperature characteristic indicates that the temperature value is always within a second temperature range, wherein the lower limit of the second temperature range is greater than the upper limit of the first temperature range.

[0097] In an exemplary embodiment, determining a thermal runaway mode based on the temperature data includes: when it is determined that the temperature data indicates that the temperature value of the vehicle battery in the current time period exhibits a fourth temperature characteristic, inputting the temperature data into a thermal runaway mode prediction model to obtain a thermal runaway mode prediction result output by the thermal runaway mode prediction model, wherein the fourth temperature characteristic is a temperature characteristic other than the first temperature characteristic, the second temperature characteristic, and the third temperature characteristic, and the thermal runaway mode prediction model is a pre-trained model used to predict the probability that the output temperature data conforms to the temperature characteristics of each thermal runaway mode; and determining the thermal runaway mode based on the thermal runaway mode with the highest probability in the thermal runaway mode prediction result.

[0098] In an exemplary embodiment, determining a target adjustment coefficient based on the thermal runaway mode includes: obtaining the current state of charge (SOC) value of the vehicle battery; determining a target mapping relationship from a plurality of mapping relationships based on the current SOC value, wherein the plurality of mapping relationships are used to record preset adjustment coefficients for different thermal runaway modes of the vehicle battery under different SOC values, and the target mapping relationship is used to record preset adjustment coefficients for the vehicle battery under the current SOC value; and determining the target adjustment coefficient based on the target mapping relationship.

[0099] In an exemplary embodiment, determining the target thermal runaway threshold based on the target adjustment coefficient and the preset thermal runaway threshold includes: obtaining a first adjustment coefficient corresponding to the mechanical runaway mode, a second adjustment coefficient corresponding to the electrical runaway mode, and a third adjustment coefficient corresponding to the high-temperature runaway mode from the target mapping relationship; if the thermal runaway mode is determined to be the mechanical runaway mode, determining the first adjustment coefficient as the target adjustment coefficient; or, if the thermal runaway mode is determined to be the electrical runaway mode, determining the second adjustment coefficient as the target adjustment coefficient; or, if the thermal runaway mode is determined to be the high-temperature runaway mode, determining the third adjustment coefficient as the target adjustment coefficient; and determining the target thermal runaway threshold based on the product of the target adjustment coefficient and the preset thermal runaway threshold.

[0100] In an exemplary embodiment, determining a target adjustment coefficient based on the thermal runaway mode includes: determining the thermal runaway mode as a multidimensional thermal runaway mode when the thermal runaway mode prediction result indicates that the probability of at least two thermal runaway modes is greater than a preset probability; and determining the target adjustment coefficient based on the probability of each thermal runaway mode among the at least two thermal runaway modes and the adjustment coefficient corresponding to each thermal runaway mode.

[0101] In one exemplary embodiment, determining the target adjustment coefficient based on the probability of each thermal runaway mode among the at least two thermal runaway modes and the adjustment coefficient corresponding to each thermal runaway mode includes: using the probability of each thermal runaway mode as a weighting coefficient, and performing a weighted summation of the adjustment coefficients corresponding to each thermal runaway mode to obtain the target adjustment coefficient.

[0102] In an exemplary embodiment, before determining the target mapping relationship from multiple mapping relationships based on the current state of charge value, the method further includes: acquiring experimental data of thermal runaway experiments conducted on the vehicle battery under different states of charge, wherein the thermal runaway experiments are used to test the thermal runaway data of the vehicle battery under different thermal runaway models; and determining the preset thermal runaway threshold and the multiple mapping relationships based on the experimental data.

[0103] In an exemplary embodiment, acquiring experimental data of the vehicle battery under different states of charge for thermal runaway experiments includes: acquiring first experimental data corresponding to the mechanical runaway mode, wherein the first experimental data is thermal runaway data of the vehicle battery under different states of charge when the battery structure is mechanically damaged; acquiring second experimental data corresponding to the electrical runaway mode, wherein the second experimental data is thermal runaway data of the vehicle battery under different states of charge when electrical parameters are abnormal; acquiring third experimental data corresponding to the high-temperature runaway mode, wherein the third experimental data is thermal runaway data of the vehicle battery under different states of charge in a third temperature range; and determining the experimental data based on the first experimental data, the second experimental data, and the third experimental data.

[0104] In an exemplary embodiment, generating an early warning instruction when the temperature rise rate in a later time period is greater than the target thermal runaway threshold includes: calculating the temperature rise rate using a time sliding window in the later time period, wherein the time window size of the time sliding window is a preset duration, and the temperature rise rate is the rate of temperature increase within the time window; and generating the early warning instruction when it is determined that the temperature rise rate is always greater than the target thermal runaway threshold during the preset sliding duration of the time sliding window.

[0105] In one exemplary embodiment, the temperature sensor is positioned at the geometric center of the largest surface of the vehicle battery not covered by the heating element, and is used to collect the temperature data at a preset sampling period.

[0106] In an exemplary embodiment, the battery management system further includes a flame sensor and a smoke sensor, both of which are communicatively connected to the controller. After generating an early warning command, the method further includes: disconnecting the high-voltage circuit of the vehicle battery to stop charging or heating the vehicle battery; controlling the vehicle battery to actively discharge at a constant power through a preset discharge resistor until the voltage of the vehicle battery drops below a safe voltage threshold; and, if the flame sensor detects a fire in the installation space of the vehicle battery, and / or if the smoke sensor detects smoke in the installation space of the vehicle battery, sending a fire extinguishing command to the battery thermal management system to activate the fire extinguishing device.

[0107] In one exemplary embodiment, after generating the warning command, the method further includes: activating a high-speed data acquisition mode to record multidimensional data of the vehicle battery during thermal runaway at a preset sampling frequency; aligning the multidimensional data according to the time axis to generate a data packet containing a timestamp, a data identifier, and a data checksum; and uploading the data packet to a cloud server and simultaneously saving it to a local storage device.

[0108] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.

[0109] Embodiments of this application also provide a storage medium including a stored program, wherein the program executes any of the methods described above when it is run.

[0110] Optionally, in this embodiment, the storage medium may be configured to store program code for performing the following steps:

[0111] S1, Obtain the temperature data of the vehicle battery during the current time period;

[0112] S2, determine the thermal runaway mode based on the temperature data, the thermal runaway mode includes mechanical runaway mode, electrical runaway mode and high temperature runaway mode;

[0113] S3, determine the target adjustment coefficient according to the thermal runaway mode, and determine the target thermal runaway threshold according to the target adjustment coefficient and the preset thermal runaway threshold;

[0114] S4, if the temperature rise rate in the subsequent period exceeds the target thermal runaway threshold, a warning command is generated, which is used to instruct the user to issue a battery thermal runaway warning.

[0115] Embodiments of this application also provide an electronic device including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0116] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0117] Optionally, in this embodiment, the processor can be configured to perform the following steps via a computer program:

[0118] S1, Obtain the temperature data of the vehicle battery during the current time period;

[0119] S2, determine the thermal runaway mode based on the temperature data, the thermal runaway mode includes mechanical runaway mode, electrical runaway mode and high temperature runaway mode;

[0120] S3, determine the target adjustment coefficient according to the thermal runaway mode, and determine the target thermal runaway threshold according to the target adjustment coefficient and the preset thermal runaway threshold;

[0121] S4, if the temperature rise rate in the subsequent period exceeds the target thermal runaway threshold, a warning command is generated, which is used to instruct the user to issue a battery thermal runaway warning.

[0122] Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0123] Optionally, embodiments of this application also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0124] Optionally, embodiments of this application also provide another computer program product, including a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0125] Optionally, embodiments of this application also provide a computer program that includes computer instructions stored in a computer-readable storage medium; a processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps in any of the above method embodiments.

[0126] Optionally, embodiments of this application also provide an electric vehicle, including a battery and the battery management system described above, wherein the battery management system is used to monitor thermal runaway of the battery.

[0127] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated here.

[0128] Obviously, those skilled in the art should understand that the modules or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented here, or they can be fabricated as separate integrated circuits, or multiple modules or steps can be fabricated as a single integrated circuit. Thus, this application is not limited to any particular hardware and software combination.

[0129] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A battery management system, characterized in that, It includes a temperature sensor and a controller, wherein the temperature sensor is communicatively connected to the controller; The temperature sensor is used to acquire the temperature data of the vehicle battery during the current time period; The controller is used for: The thermal runaway mode is determined based on the temperature data, and the thermal runaway mode includes mechanical runaway mode, electrical runaway mode and high temperature runaway mode; A target adjustment coefficient is determined based on the thermal runaway mode, and a target thermal runaway threshold is determined based on the target adjustment coefficient and a preset thermal runaway threshold. If the rate of temperature rise in a later period exceeds the target thermal runaway threshold, an early warning command is generated, which is used to instruct the user to issue a battery thermal runaway warning.

2. The battery management system according to claim 1, characterized in that, The controller is also configured to determine the thermal runaway mode in the following ways: If the temperature data indicates that the temperature value of the vehicle battery in the current time period is a first temperature characteristic, the mechanical runaway mode is determined to be the thermal runaway mode, wherein the first temperature characteristic is used to indicate that the temperature value is always within a first temperature range. If the temperature data indicates that the temperature value of the vehicle battery in the current time period exhibits a second temperature characteristic, the electrical runaway mode is determined to be the thermal runaway mode, wherein the second temperature characteristic is used to indicate that the temperature value continues to rise and the temperature rise is greater than a preset temperature difference. If the temperature data indicates that the temperature value of the vehicle battery in the current time period exhibits a third temperature characteristic, the high-temperature runaway mode is determined as the thermal runaway mode, wherein the third temperature characteristic is used to indicate that the temperature value is always within a second temperature range, wherein the lower limit of the second temperature range is greater than the upper limit of the first temperature range.

3. The battery management system according to claim 2, characterized in that, The controller is also configured to determine the thermal runaway mode in the following ways: If the temperature data indicates that the temperature value of the vehicle battery in the current time period exhibits a fourth temperature characteristic, the temperature data is input into the thermal runaway mode prediction model to obtain the thermal runaway mode prediction result output by the thermal runaway mode prediction model. The fourth temperature characteristic is a temperature characteristic other than the first temperature characteristic, the second temperature characteristic, and the third temperature characteristic. The thermal runaway mode prediction model is pre-trained and used to predict the probability that the output temperature data conforms to the temperature characteristics of each thermal runaway mode. The thermal runaway mode is determined based on the thermal runaway mode with the highest probability in the predicted thermal runaway mode results.

4. The battery management system according to claim 3, characterized in that, The controller is also configured to determine the target adjustment coefficient in the following manner: Obtain the current state of charge value of the vehicle battery; A target mapping relationship is determined from multiple mapping relationships based on the current state of charge value. The multiple mapping relationships are used to record the preset adjustment coefficients for different thermal runaway modes of the vehicle battery under different states of charge. The target mapping relationship is used to record the preset adjustment coefficients for the vehicle battery under the current state of charge value. The target adjustment coefficient is determined based on the target mapping relationship.

5. The battery management system according to claim 4, characterized in that, The controller is also configured to determine the target thermal runaway threshold in the following manner: The first adjustment coefficient corresponding to the mechanical runaway mode, the second adjustment coefficient corresponding to the electrical runaway mode, and the third adjustment coefficient corresponding to the high-temperature runaway mode are obtained from the target mapping relationship. If the thermal runaway mode is determined to be the mechanical runaway mode, the first adjustment coefficient is determined as the target adjustment coefficient; Alternatively, if the thermal runaway mode is determined to be the electrical runaway mode, the second adjustment coefficient is determined as the target adjustment coefficient; Alternatively, if the thermal runaway mode is determined to be the high-temperature runaway mode, the third adjustment coefficient is determined as the target adjustment coefficient; The target thermal runaway threshold is determined by multiplying the target adjustment coefficient by the preset thermal runaway threshold.

6. The battery management system according to claim 5, characterized in that, The controller is also configured to determine the target adjustment coefficient in the following manner: If the thermal runaway mode prediction results indicate that the probability of at least two thermal runaway modes is greater than a preset probability, the thermal runaway mode is determined to be a multidimensional thermal runaway mode. The target adjustment coefficient is determined based on the probability of each thermal runaway mode among the at least two thermal runaway modes and the adjustment coefficient corresponding to each thermal runaway mode.

7. The battery management system according to claim 6, characterized in that, The controller is also configured to determine the target adjustment coefficient in the following manner: The probability of each thermal runaway mode is used as a weighting coefficient, and the adjustment coefficients corresponding to each thermal runaway mode are summed in a weighted manner to obtain the target adjustment coefficient.

8. The battery management system according to claim 4, characterized in that, It also includes a data acquisition module, which is used for: The experimental data of thermal runaway experiments of the vehicle battery under different states of charge are obtained, wherein the thermal runaway experiments are used to test the thermal runaway data of the vehicle battery under different thermal runaway models. The preset thermal runaway threshold and the multiple mapping relationships are determined based on the experimental data.

9. The battery management system according to claim 8, characterized in that, The data acquisition module is also used to acquire the experimental data in the following ways: Obtain the first experimental data corresponding to the mechanical runaway mode, wherein the first experimental data is the thermal runaway data of the vehicle battery under the condition that the battery structure is subjected to mechanical damage under different states of charge; Obtain the second experimental data corresponding to the electrical runaway mode, wherein the second experimental data is the thermal runaway data of the vehicle battery under the abnormal electrical parameters of different states of charge; Obtain the third experimental data corresponding to the high temperature runaway mode, wherein the third experimental data is the thermal runaway data of the vehicle battery under the different states of charge in the third temperature range; The experimental data are determined based on the first experimental data, the second experimental data, and the third experimental data.

10. The battery management system according to claim 1, characterized in that, The controller is also configured to generate the warning instruction in the following manner: In the latter period, the temperature rise rate is calculated using a time sliding window, wherein the size of the time sliding window is a preset duration, and the temperature rise rate is the rate of temperature increase within the time window. If the temperature rise rate is always greater than the target thermal runaway threshold during the preset sliding window period, the warning command is generated.

11. The battery management system according to claim 1, characterized in that, The temperature sensor is located at the geometric center of the largest surface of the vehicle battery not covered by the heating element, and is used to collect the temperature data at a preset sampling period.

12. The battery management system according to claim 1, characterized in that, It also includes a flame sensor and a smoke sensor, both of which are communicatively connected to the controller. After generating the warning command, the controller is further configured to: Disconnect the high-voltage circuit of the vehicle battery to stop charging or heating the vehicle battery; The vehicle battery is controlled to actively discharge at a constant power through a preset discharge resistor until the voltage of the vehicle battery drops below a safe voltage threshold. If the flame sensor detects a fire in the mounting space of the vehicle battery, and / or if the smoke sensor detects smoke in the mounting space of the vehicle battery, a fire extinguishing command is sent to the battery thermal management system to activate the fire extinguishing device.

13. The battery management system according to claim 1, characterized in that, After generating the warning instruction, the controller is further configured to: Start the high-speed data acquisition mode to record multi-dimensional data of the vehicle battery during thermal runaway at a preset sampling frequency; The multidimensional data is aligned according to the time axis to generate a data packet containing timestamps, data identifiers, and data check codes; The data packet is uploaded to the cloud server and simultaneously saved to the local storage.

14. A method for early warning of battery thermal runaway, characterized in that, Applications in battery management systems include: Obtain the temperature data of the vehicle battery during the current time period; The thermal runaway mode is determined based on the temperature data, and the thermal runaway mode includes mechanical runaway mode, electrical runaway mode and high temperature runaway mode; A target adjustment coefficient is determined based on the thermal runaway mode, and a target thermal runaway threshold is determined based on the target adjustment coefficient and a preset thermal runaway threshold. If the rate of temperature rise in a later period exceeds the target thermal runaway threshold, an early warning command is generated, which is used to instruct the user to issue a battery thermal runaway warning.

15. An electric vehicle, characterized in that, The battery includes a battery and a battery management system according to any one of claims 1-13, the battery management system being used for thermal runaway monitoring of the battery.

Images