Thermal runaway early warning protection method and energy storage system

Abstract

This application discloses a thermal runaway early warning and protection method for an energy storage system. The method includes: acquiring the cell temperature, charging / discharging current, and ambient temperature of the battery pack of the energy storage system; calculating the theoretical temperature rise rate of the cell based on the cell temperature, the charging / discharging current, the ambient temperature, and the heat generation power, heat dissipation power, and heat capacity parameters of the cell in the battery pack; and controlling the energy storage system to perform an early warning and protection action when the ratio of the actual temperature rise rate of the cell to the theoretical temperature rise rate exceeds a predetermined threshold. Thus, the thermal runaway early warning and protection method does not rely on a fixed temperature threshold or a fixed temperature rise rate threshold, but adaptively determines whether the temperature rise is abnormal based on the current operating conditions, effectively avoiding the problems of false alarms under high current conditions and missed alarms under low current conditions caused by a fixed temperature rise rate threshold. This application also discloses an energy storage system.

Description

Technical Field

[0001] This invention relates to the field of vehicle technology, and in particular to a thermal runaway early warning and protection method and an energy storage system. Background Technology

[0002] Battery management systems (BMS) in energy storage systems prevent thermal runaway by monitoring the cell temperature of the battery pack. For example, a temperature threshold of 60°C is set. When the temperature sensor detects that the cell temperature has reached the threshold, the BMS triggers overheat protection and disconnects the battery pack's charging and discharging circuits. However, in practical applications, when the BMS triggers overheat protection, the battery cells are often already close to the thermal runaway threshold, leaving very little time for personnel to handle the situation or evacuate, making early warning impossible.

[0003] Therefore, related technologies add temperature rise rate monitoring to the existing temperature threshold. For example, the temperature rise rate threshold is 5℃ / min. When the cell temperature rise rate exceeds the threshold, the battery management system triggers an overheat warning or overheat protection. However, the temperature rise rate of a cell varies greatly at different charge / discharge rates and ambient temperatures: the temperature rise rate may reach 4-5℃ / min during high-current fast charging, while it may be less than 0.5℃ / min during low-current slow charging. Therefore, if the temperature rise threshold is set too high, abnormal temperature rises under low-current conditions (e.g., 2℃ / min) may be missed; if the threshold is set too low, normal temperature rises under high-current conditions will frequently trigger false alarms. Summary of the Invention

[0004] This application provides a battery thermal runaway early warning system and an energy storage system.

[0005] This application provides a thermal runaway early warning and protection method for an energy storage system, the method comprising: The cell temperature, charging and discharging current, and ambient temperature of the battery pack of the energy storage system are obtained. Based on the cell temperature, the charging / discharging current, the ambient temperature, and the heat generation, heat dissipation, and heat capacity parameters of the cells in the battery pack, the theoretical temperature rise rate of the cell is calculated; and If the ratio of the actual temperature rise rate of the battery cell to the theoretical temperature rise rate exceeds a predetermined threshold, the energy storage system is controlled to perform an early warning protection action.

[0006] The thermal runaway early warning protection method of this application acquires the cell temperature, charging and discharging current, and ambient temperature of the battery pack in the energy storage system. Based on the cell's heat generation power, heat dissipation power, and heat capacity parameters, it dynamically calculates the theoretical temperature rise rate of the cell. When the ratio of the actual temperature rise rate to the theoretical temperature rise rate exceeds a predetermined threshold, it controls the energy storage system to execute an early warning protection action. Thus, the thermal runaway early warning protection method does not rely on a fixed temperature threshold or a fixed temperature rise rate threshold, but rather adaptively determines whether the temperature rise is abnormal based on the current operating conditions.

[0007] Specifically, when the battery pack is fast-charged with a high current, the heat generation is large, and the theoretical temperature rise rate is inherently high. Even if the actual temperature rise rate reaches 4 to 5 degrees Celsius per minute, it may still be within the normal ratio range and will not trigger a false alarm. When the battery pack is slow-charged with a low current, the theoretical temperature rise rate is low, and the actual temperature rise rate of 2 degrees Celsius per minute may cause the ratio to exceed the predetermined threshold. The thermal runaway early warning protection method can accurately identify the abnormality and issue an early warning. It unifies the sensitivity to abnormalities under different charge and discharge rates and ambient temperatures, and effectively avoids the problems of false alarms under high current conditions and missed alarms under low current conditions caused by fixed temperature rise rate thresholds.

[0008] Meanwhile, since the theoretical temperature rise rate is calculated based on a physical model of heat generation power, heat dissipation power, and heat capacity parameters, it can reflect the current heat generation and heat dissipation balance of the battery cell. When a micro-short circuit occurs inside the battery cell, the internal resistance increases abnormally, or the heat dissipation conditions deteriorate, the actual temperature rise rate will deviate significantly from the theoretical value. The thermal runaway early warning protection method can issue an early warning before the battery cell temperature reaches the traditional protection threshold (e.g., 60°C), buying valuable time for personnel to handle the situation and evacuate. In addition, the thermal runaway early warning protection method does not rely on the absolute temperature of the battery cell and can effectively identify abnormal temperature rises even in low-temperature environments, thereby improving the overall safety of the energy storage system.

[0009] In some embodiments, calculating the theoretical temperature rise rate of the battery cell includes: Calculate the heat generation power based on the charging and discharging current and the internal resistance of the battery cell; Calculate the heat dissipation power based on the cell temperature, the ambient temperature, the cell's heat dissipation coefficient, and the heat dissipation area; and The theoretical temperature rise rate is calculated based on the heat generation power, the heat dissipation power, and the thermal capacity parameters of the battery cell.

[0010] The thermal runaway early warning and protection method of this application calculates the heat generation power through internal resistance and current, and the heat dissipation power through temperature difference, heat dissipation coefficient, and heat dissipation area. It then combines these with heat capacity to derive the theoretical temperature rise rate. Thus, the theoretical temperature rise rate reflects the dynamic balance between Joule heating and heat dissipation, allowing the reference value to change in real time with charging / discharging current, ambient temperature, and cell temperature. Compared to the fixed lookup table method, this method better reflects the actual physical process, thereby significantly improving the accuracy of anomaly detection.

[0011] In some embodiments, calculating the theoretical temperature rise rate of the battery cell before calculating the heat generation power further includes: The internal resistance of the battery cell is updated based on the cumulative charge / discharge amount or operating time of the battery pack.

[0012] The thermal runaway early warning protection method of this application can update the internal resistance of the battery cell online or offline. Since the internal resistance of the battery cell changes with aging, temperature, and state of charge, a fixed factory-set internal resistance can lead to deviations in the calculation of heat generation power. By updating the internal resistance periodically or as needed, the theoretical temperature rise rate model can adapt to the health status of the battery cell, further reducing false alarms and missed alarms, and extending the effective service life of the thermal runaway early warning protection method.

[0013] In some embodiments, the predetermined threshold includes a first threshold and a second threshold, wherein the second threshold is greater than the first threshold; controlling the energy storage system to perform an early warning protection action includes: When the ratio is greater than or equal to the first threshold but less than the second threshold, the energy storage system is controlled to perform an early warning action. When the ratio is greater than or equal to the second threshold, the energy storage system is controlled to perform a protection action.

[0014] The thermal runaway early warning protection method of this application sets two threshold levels, which trigger different levels of response. When the ratio is greater than or equal to the first threshold but less than the second threshold, it indicates a minor abnormality in the battery cell. In this case, only an early warning action is performed, such as reducing the charging and discharging power or issuing an audible and visual alarm, which alerts the user without forcibly shutting down the device, balancing safety and user experience. When the ratio is greater than or equal to the second threshold, it indicates a serious abnormality in the battery cell. In this case, a protection action is performed, such as cutting off the charging and discharging circuit and forcibly isolating the fault, avoiding the user inconvenience caused by a one-size-fits-all approach, while providing the strongest protection in truly dangerous situations.

[0015] In some implementations, the warning action includes reducing the charging and discharging power of the battery pack and / or issuing an audible and visual alarm signal; the protection action includes disconnecting the charging and discharging circuit of the battery pack.

[0016] The thermal runaway early warning and protection method of this application reduces the charging and discharging power when an early warning is issued, which can slow down the deterioration rate of abnormal temperature rise. At the same time, an audible and visual alarm prompts on-site personnel to pay attention to the abnormal state. When protection is issued, the charging and discharging circuit is directly cut off, fundamentally eliminating the energy source of thermal runaway. This method can effectively intervene in the abnormal development without affecting the normal use of the energy storage system due to over-protection.

[0017] In some embodiments, after controlling the energy storage system to perform an early warning action, the step of controlling the energy storage system to perform an early warning protection action further includes: When the ratio drops below a third threshold, the energy storage system is controlled to stop the warning action; the third threshold is less than the first threshold.

[0018] The thermal runaway early warning protection method of this application incorporates an automatic recovery mechanism with hysteresis. When the abnormal temperature rise subsides and the ratio of the actual temperature rise rate to the theoretical temperature rise rate falls back to a third threshold (e.g., 1.3 times) below the first threshold, the early warning action is automatically lifted, allowing the energy storage system to return to normal operation. This avoids alarm jitter caused by frequent triggering and lifting of early warnings near the critical value, and requires no manual intervention, thus improving the level of intelligence and user experience.

[0019] In some embodiments, after controlling the energy storage system to perform a protection action, the step of controlling the energy storage system to perform an early warning protection action further includes: Locking the protection action; and The protection action is released upon receiving an external manual reset signal.

[0020] The thermal runaway early warning and protection method of this application adopts a latching mechanism for protection actions, which requires manual inspection and troubleshooting before manual reset. This prevents the energy storage system from automatically restarting when the fault is not eliminated, avoids repeated risks, complies with the requirement in safety specifications that serious faults must be manually confirmed, and ensures the safe use of the energy storage system.

[0021] In some embodiments, controlling the energy storage system to perform early warning and protection actions further includes: The cell temperature, the charging and discharging current, and the ratio are stored in the storage module of the energy storage system.

[0022] The thermal runaway early warning and protection method of this application stores key data at abnormal moments in the storage module, providing objective evidence for post-fault analysis. This helps to determine the cause of the abnormality (e.g., whether it is caused by increased internal resistance, poor heat dissipation or micro-short circuit), and also facilitates manufacturers to optimize algorithms or users to trace the accident process, thereby improving the maintainability and accident analysis capabilities of the energy storage system.

[0023] In some embodiments, controlling the energy storage system to perform early warning and protection actions further includes: The communication module of the energy storage system is controlled to send early warning and protection information to the remote monitoring platform or user terminal.

[0024] The thermal runaway early warning and protection method of this application pushes early warning and protection information to the cloud or user's mobile phone in real time through a communication module (such as CAN bus or 4G module), so that unattended energy storage systems can also know the battery status in a timely manner, which facilitates rapid response and prevents the accident from escalating. It is especially suitable for distributed energy storage, residential energy storage and other scenarios, and improves the remote management capability of energy storage systems.

[0025] In some embodiments, the energy storage system includes multiple cell temperature probes for detecting the temperature of multiple cells at multiple locations within the battery pack. The acquisition of the cell temperature, charging / discharging current, and ambient temperature of the battery pack of the energy storage system includes: Obtain the temperature of the multiple battery cells; The calculation of the theoretical temperature rise rate of the battery cell includes: Calculate the theoretical temperature rise rates corresponding to the temperatures of the multiple battery cells; The control of the energy storage system to perform early warning and protection actions includes: Calculate multiple ratios between multiple actual temperature rise rates and multiple theoretical temperature rise rates corresponding to multiple cell temperatures; and If any of the ratios exceeds the predetermined threshold, the energy storage system is controlled to perform the early warning protection action.

[0026] The thermal runaway early warning method of this application sets cell temperature probes at multiple locations within the battery pack to acquire multiple cell temperatures, and calculates the theoretical temperature rise rate and actual temperature rise rate for each location, obtaining multiple ratios. When any ratio exceeds a predetermined threshold, the method immediately triggers an early warning. Since temperatures may vary at different locations within the battery pack (e.g., localized cell aging, micro-short circuits, or uneven heat dissipation), monitoring only a single temperature point may not promptly detect localized anomalies. By calculating the theoretical temperature rise rate for multiple locations and comparing the actual temperature rise rate, the thermal runaway early warning method can more sensitively identify abnormal temperature rises at any location within the battery pack, preventing the spread of thermal runaway due to undetected localized overheating. Simultaneously, the theoretical temperature rise rate for each location is dynamically calculated based on the current heat generation and dissipation conditions at that location, adapting to different operating conditions and effectively avoiding missed or false alarms caused by fixed thresholds, further improving the safety and reliability of the energy storage system.

[0027] This application also provides an energy storage system, the energy storage system comprising: Battery pack; The data acquisition module is used to collect data on the cell temperature, charging / discharging current, and ambient temperature of the battery pack; and The main control module is used to execute the thermal runaway early warning and protection method. The early warning and protection module is used to execute the aforementioned early warning and protection actions. Attached Figure Description

[0028] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, wherein: Figure 1 This is a flowchart illustrating a thermal runaway early warning and protection method according to certain embodiments of this application; Figure 2 This is a schematic diagram of the functional modules of an energy storage system according to certain embodiments of this application; Figure 3-9 This is a flowchart illustrating a thermal runaway early warning and protection method according to certain embodiments of this application. Detailed Implementation

[0029] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the embodiments of this application, and should not be construed as limiting the embodiments of this application.

[0030] The thermal runaway early warning and protection method provided in this application is applicable to various energy storage systems that use lithium-ion batteries, lithium iron phosphate batteries or other electrochemical energy storage media, including but not limited to: residential energy storage, industrial and commercial energy storage, backup power for communication base stations, grid-side energy storage power stations and mobile energy storage devices.

[0031] Energy storage systems typically include a battery pack and a battery management system (BMS).

[0032] A battery pack consists of multiple cells connected in series or parallel. A cell is the smallest independent unit where an electrochemical reaction occurs, and its internal structure includes a positive electrode, a negative electrode, a separator, and an electrolyte. For example, a typical residential energy storage battery pack may consist of 16 lithium iron phosphate cells connected in series, with a nominal voltage of 48V and a capacity of 100Ah; a commercial or industrial energy storage battery pack may consist of 240 cells first connected in parallel and then in series, forming an 800V / 280Ah battery cluster. During the operation of an energy storage system, thermal runaway is the most serious failure mode threatening the safety of the system. Thermal runaway is usually induced by mechanical abuse (such as puncture, crushing), electrical abuse (such as overcharging, internal short circuits), or thermal abuse (such as high-temperature environments). When a local short circuit or side reaction occurs inside the cell, the rate of heat generation exceeds the rate of heat dissipation, causing a rapid rise in temperature. This triggers a series of chain exothermic reactions, such as separator shrinkage, positive electrode decomposition, and electrolyte combustion, ultimately leading to fire or even explosion. Therefore, timely and accurate identification of early signs of thermal runaway is crucial for ensuring the safety of energy storage systems.

[0033] In related technologies, the Battery Management System (BMS) prevents thermal runaway of the energy storage system by monitoring the cell temperature of the battery pack. For example, a temperature threshold of 60°C is set. When the temperature sensor of the energy storage system detects that the cell temperature has reached the threshold, the BMS triggers overheat protection and cuts off the charging and discharging circuit of the battery pack. However, in actual applications, when the BMS triggers overheat protection, the battery cells in the battery pack are often already close to the thermal runaway critical point, leaving very little time for personnel to handle or evacuate, making early warning impossible.

[0034] Therefore, related technologies add monitoring of the temperature rise rate to the temperature threshold, for example, a temperature rise rate threshold of 5℃ / min. When the cell temperature rise rate exceeds the temperature rise rate threshold, the battery management system triggers an overheat warning or overheat protection. However, in actual operation, the battery pack may experience fast charging (e.g., drawing power from the grid to cope with upcoming peak electricity demand), slow charging (e.g., supplementing photovoltaic surplus power), high-current discharging (e.g., charging air conditioners or electric vehicles), or low-current discharging (e.g., standby lighting at night). The temperature rise characteristics of the cells differ significantly under different operating conditions. The temperature rise rate during high-current fast charging may reach 4-5℃ / min, while the temperature rise during low-current slow charging may be less than 0.5℃ / min. Therefore, if the temperature rise threshold is set too high, abnormal temperature rises under low-current conditions (e.g., 2℃ / min) may be missed; if the threshold is set too low, normal temperature rises under high-current conditions will frequently trigger false alarms.

[0035] In view of this, please refer to Figure 1 This application provides a thermal runaway early warning and protection method for an energy storage system, the method comprising: S01: Obtain the cell temperature, charging and discharging current, and ambient temperature of the battery pack in the energy storage system; S02: Calculate the theoretical temperature rise rate of the battery cell based on cell temperature, charging / discharging current, ambient temperature, and the cell's heat generation power, heat dissipation power, and heat capacity parameters within the battery pack; and S03: When the ratio of the actual temperature rise rate of the battery cell to the theoretical temperature rise rate exceeds a predetermined threshold, the energy storage system is controlled to perform an early warning protection action.

[0036] Please see Figure 2 This application also provides an energy storage system 10. The energy storage system 10 includes a battery pack 12, a data acquisition module 14, a main control module 16, and a warning and protection module 18. The data acquisition module 14 is used to acquire the cell temperature, charging and discharging current, and ambient temperature of the battery pack 12. The main control module 16 is used to execute steps S01-S03. The warning and protection module 18 is used to execute warning and protection actions. That is to say, the thermal runaway warning and protection method of this application embodiment can be applied to the energy storage system 10.

[0037] Specifically, in step S01, the cell temperature refers to the temperature value measured by a temperature sensor attached to the cell surface (usually chosen at the geometric center of the large surface area of ​​the cell or near the positive and negative tabs). The temperature sensor can be a negative temperature coefficient thermistor, whose resistance decreases as temperature increases, and is converted into a digital temperature value through a voltage divider circuit and an analog-to-digital converter (ADC). Another option is to use a thermocouple (e.g., a type K thermocouple), whose output is a millivolt-level differential voltage, requiring cold junction compensation and a dedicated amplifier. In applications requiring high accuracy, a digital temperature chip can be used, directly reading the temperature value via a single bus or I2C interface, eliminating the need for calibration. For multiple cells within a battery pack, multiple temperature sensors can be arranged, for example, one NTC for every four cells, or one each in the most heat-prone central area and near the tabs. The sampling period for cell temperature is typically 100ms to 1s, which can be adjusted according to operating conditions: shortened to 100ms when current changes drastically or temperature rises rapidly; extended to 1s during steady-state operation to reduce computational load.

[0038] The charging and discharging current refers to the DC current flowing through the main circuit of the battery pack (12), with the direction being positive for charging and negative for discharging. Current measurement typically employs two methods: one is a shunt (low-resistance precision resistor) connected in series at the negative terminal of the main circuit, with the millivolt-level voltage across its terminals amplified by an isolation amplifier before being fed into the ADC; the other is a Hall effect current sensor, which utilizes the principle of magnetic field induction to output a voltage proportional to the current, offering advantages such as electrical isolation and no insertion loss. For high-current energy storage systems (hundreds to thousands of amperes), fluxgate current sensors or Rogowski coils are commonly used; the former offers high accuracy, while the latter provides good linearity and avoids magnetic saturation issues. When calculating heat generation power, only the absolute value of the current is used because Joule heating is independent of the current direction.

[0039] Ambient temperature refers to the temperature of the external environment where the battery pack 12 is located, typically obtained through a temperature sensor placed at the air inlet of the battery pack 12 casing (at least 10cm away from the battery pack) or at the air conditioner return vent. Ambient temperature is used to calculate heat dissipation power and is an important parameter for evaluating heat dissipation conditions. When the battery pack is installed in a temperature-controlled computer room, the ambient temperature can be considered a constant value (e.g., 25°C), or the temperature sensor value of the air conditioning system can be read directly, or the local temperature value can be read from the cloud.

[0040] In step S02, the theoretical temperature rise rate is physically defined as: given a charge / discharge current, current cell temperature, and ambient temperature, if the battery is only affected by Joule heating and heat dissipation (ignoring other side reaction heat), the expected change in its temperature per unit time. It is a dynamic benchmark that changes in real time with operating conditions. Calculating the theoretical temperature rise rate requires the cell's heat generation power, heat dissipation power, and heat capacity parameters.

[0041] The heat production power originates from Joule heat, and its formula is Q_gen=I 2 ×R, where I is the absolute value of the charging / discharging current (in A), and R is the DC internal resistance of the battery cell (in Ω). The internal resistance R is not constant; it varies with cell temperature, state of charge (SOC), and aging. For example, a lithium iron phosphate cell with a nominal internal resistance of 0.5 mΩ has a measured internal resistance of approximately 0.5 mΩ at 25°C and 50% SOC, but its internal resistance may rise to 1.0 mΩ at 0°C and drop to 0.45 mΩ at 100% SOC. To obtain a more accurate heat generation power, a two-dimensional lookup table of internal resistance variations with temperature and SOC can be established experimentally beforehand. In step S022, the internal resistance value is obtained by interpolation based on the current temperature and SOC.

[0042] The formula for calculating heat dissipation power is Q_diss=k×A×(T_cell-T_amb), where k is the heat dissipation coefficient (unit: W / (m²·K)), A is the effective heat dissipation area of ​​the cell (unit: m²), T_cell is the current cell temperature, and T_amb is the ambient temperature. The heat dissipation coefficient k is closely related to the cooling method of the battery pack 12: under natural cooling, the air convection coefficient is about 5-10 W / (m²·K); under forced air cooling, the fan can increase k to 15-30 W / (m²·K); under liquid cooling, using ethylene glycol aqueous solution circulation, k can reach 50-100 W / (m²·K). The heat dissipation area A refers to the effective area of ​​the cell surface in contact with the cooling medium. For prismatic cells, it is usually taken as the total area of ​​the six sides of the cell minus the part blocked by adjacent cells. For example, a prismatic lithium iron phosphate cell with dimensions of 173×125×45mm has a total surface area of ​​approximately 0.06m², but considering the close arrangement of the cells, the actual effective heat dissipation area may only be 0.02m². The heat capacity parameter C (unit: J / K) represents the amount of heat required to raise the cell temperature by 1℃, and can be obtained through calorimetry experiments. The heat capacity of a common prismatic lithium iron phosphate cell (capacity 100Ah) is approximately 800-1200J / K, while that of a cylindrical cell (18650) is approximately 50-80J / K.

[0043] The final formula for calculating the theoretical temperature rise rate is: ΔT_theory=(Q_gen-Q_diss) / C×Δt, where Δt is the calculation period (usually 1 second). If the calculated ΔT_theory≤0 (i.e., heat generation is less than heat dissipation, and cooling is expected), to avoid division by zero or negative ratios in subsequent ratio calculations, the theoretical temperature rise rate is set to a very small positive number (e.g., 0.001℃ / min) in actual programming, while limiting the heat dissipation power to no more than the heat generation power. Another approach is to directly determine it as normal (without triggering a warning) when ΔT_theory≤0, because thermal runaway cannot occur under cooling conditions.

[0044] For heat generation, polarization heat and reaction heat can also be considered. Polarization heat is caused by overpotential in the electrochemical reaction and is related to current and polarization resistance, which varies with current density. In high-rate pulse charge / discharge scenarios, polarization heat may become the main heat source. In this case, an equivalent circuit model (such as the PNGV model or Thevenin model) can be used to calculate the total internal resistance, including ohmic resistance and polarization resistance. For heat dissipation, if there are heat conduction paths inside the battery pack (such as thermally conductive silicone pads between cells), a heat conduction model needs to be introduced, using the finite element method or lumped parameter thermal network method. For simplification, this implementation uses a lumped parameter thermal model, treating the entire cell as an isothermal body, which is sufficient for most application scenarios.

[0045] In step S03, the actual temperature rise rate ΔT_actual is obtained by differential calculation of continuously sampled cell temperature data. The simplest calculation method is the endpoint differential method: select the starting temperature T_start and the ending temperature T_current within a time window (e.g., 60 seconds), and calculate ΔT_actual = (T_start - T_current) / Δt. The advantage of this method is its low computational cost, but it is sensitive to noise. A more robust method is the linear regression method: collect m temperature-time points (t_i, T_i) within the time window, and fit a straight line T = a·t + b using the least squares method; the slope a is the temperature rise rate. This method can effectively suppress sampling noise. The length of the time window can be selected according to the system response requirements: a 30-second window has a fast response but is easily affected by fluctuations, while a 120-second window is smooth but has a large warning delay. In this embodiment, a 60-second window is used to balance response speed and stability. The ratio of the actual temperature rise rate to the theoretical temperature rise rate, K = ΔT_actual / ΔT_theory, is dimensionless. When K≈1, it indicates that the actual temperature rise is consistent with the theoretical expectation, and the battery is in normal condition. When K>1, it indicates that the actual temperature rise is higher than expected, and there may be an anomaly. When K<1, it indicates that the actual temperature rise is lower than expected, which may be due to reduced current or improved heat dissipation, and is usually not considered an anomaly. The predetermined threshold is set based on a large amount of experimental data. For example, charge and discharge tests are performed on normal battery packs under different operating conditions, and K values ​​are collected. The 99th percentile is taken as the first threshold (e.g., 1.5), and the 99.9th percentile is taken as the second threshold (e.g., 2.0). Alternatively, the change curve of K value can be obtained through fault simulation experiments (e.g., nail penetration, overcharge), and the critical value of K before thermal runaway occurs can be selected as the threshold.

[0046] Assume a battery pack is being charged at a constant current of 0.5C (50A), with an ambient temperature of 25℃ and an initial cell temperature of 30℃. Internal resistance R = 0.5mΩ, kA = 10W / ℃, and C = 1000J / K. The theoretical temperature rise rate is calculated as: Q_gen = 50² × 0.0005 = 1.25W. The average temperature difference is taken as (30 + 33) / 2 - 25 = 6.5℃, Q_diss = 10 × 6.5 = 65W. The net heat generation is negative, and the theoretical temperature rise rate should be negative, but the actual temperature rise is 3℃ / min. This indicates an error in the model parameters, requiring calibration of kA or R. After calibration (e.g., adjusting kA = 2W / ℃), Q_diss = 2 × 6.5 = 13W, and the net heat generation = 1.25 - 13 = -11.75W, still negative. This indicates that the model predicted a temperature decrease while the actual temperature rise is negative, requiring further adjustment. This example illustrates that model parameters must be iteratively optimized using a large amount of actual operating data. To obtain a reasonable example, liquid cooling system parameters are used: kA = 100W / ℃, then Q_diss = 100 × 6.5 = 650W, which is far greater than the heat generated, and is obviously unreasonable. The correct parameters should make ΔT_theory and ΔT_actual basically consistent under normal operating conditions. Assuming that after calibration, at 50A charging, kA = 5W / ℃, then Q_diss = 5 × 6.5 = 32.5W, which is still greater than 1.25W. In reality, in addition to Joule heat, there is also reaction heat (endothermic or exothermic) during cell charging. The reaction heat during lithium iron phosphate charging is slightly endothermic, so the net heat generated may be less than the Joule heat. To simplify, the heat dissipation coefficient in this application can be adjusted to match the theoretical temperature rise with the actual temperature rise. For example, if the actual temperature rise rate is 1.5℃ / min, the required net heat production is C×ΔT_actual=1000×1.5 / 60=25W. Therefore, the model parameters need to be set so that Q_gen-Q_diss=25W. This can be achieved by adjusting kA or the internal resistance R. In actual products, the model parameters are calibrated through temperature rise experiments conducted in the hot chamber before shipment, and are not arbitrarily assumed.

[0047] The ratio K can be calculated using the difference form: (ΔT_actual - ΔT_theory) / ΔT_theory × 100%, yielding the relative deviation percentage. The corresponding thresholds can be set to 50% and 100%. This method provides a more intuitive physical meaning, but the calculation is slightly more complex. Both methods are equivalent; this application uses the ratio form, which is simpler.

[0048] In some embodiments, the acquisition module 14 may include a cell temperature acquisition module 142, a current acquisition module 144, a voltage acquisition module 146, and an ambient temperature acquisition module 148, which are used to acquire the cell temperature, charging and discharging current, cell voltage, and ambient temperature of the battery pack 12, respectively.

[0049] Specifically, before step S01, the cell temperature acquisition module 14 acquires the cell temperature of the battery pack 12, the current acquisition module 144 (e.g., a shunt or Hall current sensor connected in series in the main circuit) acquires the charging and discharging current at a fixed period (e.g., 100ms), and the voltage acquisition module 146 (e.g., an analog front-end AFE chip) simultaneously acquires the cell voltage for other BMS functions; the ambient temperature acquisition module 148 (e.g., a digital temperature sensor arranged at the air inlet of the battery pack 12) acquires the ambient temperature. In step S01, the main control module 16 acquires this data via an internal bus (e.g., SPI, CAN).

[0050] In step S02, the main control module 16 calculates the theoretical temperature rise rate based on the received cell temperature, charging / discharging current, and ambient temperature, using the cell's heat generation power, heat dissipation power, and heat capacity parameters. The main control module 16 internally stores model parameters such as the cell's internal resistance, heat dissipation coefficient, heat dissipation area, and heat capacity.

[0051] In step S03, when the ratio exceeds a predetermined threshold, the main control module 16 controls the early warning protection module 18 to perform an early warning protection action.

[0052] The thermal runaway early warning protection method of this application acquires the cell temperature, charging and discharging current, and ambient temperature of the battery pack 12 of the energy storage system 10. Based on the cell's heat generation power, heat dissipation power, and heat capacity parameters, it dynamically calculates the theoretical temperature rise rate of the cell. When the ratio of the actual temperature rise rate to the theoretical temperature rise rate exceeds a predetermined threshold, it controls the energy storage system to execute an early warning protection action. Thus, the thermal runaway early warning protection method does not rely on a fixed temperature threshold or a fixed temperature rise rate threshold, but rather adaptively determines whether the temperature rise is abnormal based on the current operating conditions.

[0053] Specifically, when the battery pack is fast-charged with a high current, the heat generation is large, and the theoretical temperature rise rate is inherently high. Even if the actual temperature rise rate reaches 4 to 5 degrees Celsius per minute, it may still be within the normal ratio range and will not trigger a false alarm. When the battery pack is slow-charged with a low current, the theoretical temperature rise rate is low, and the actual temperature rise rate of 2 degrees Celsius per minute may cause the ratio to exceed the predetermined threshold. The thermal runaway early warning protection method can accurately identify the abnormality and issue an early warning. It unifies the sensitivity to abnormalities under different charge and discharge rates and ambient temperatures, and effectively avoids the problems of false alarms under high current conditions and missed alarms under low current conditions caused by fixed temperature rise rate thresholds.

[0054] Meanwhile, since the theoretical temperature rise rate is calculated based on a physical model of heat generation power, heat dissipation power, and heat capacity parameters, it can reflect the current heat generation and heat dissipation balance of the battery cell. When a micro-short circuit occurs inside the battery cell, the internal resistance increases abnormally, or the heat dissipation conditions deteriorate, the actual temperature rise rate will deviate significantly from the theoretical value. The thermal runaway early warning protection method can issue an early warning before the battery cell temperature reaches the traditional protection threshold (e.g., 60°C), buying valuable time for personnel to handle the situation and evacuate. In addition, the thermal runaway early warning protection method does not rely on the absolute temperature of the battery cell and can effectively identify abnormal temperature rises even in low-temperature environments, thereby improving the overall safety of the energy storage system.

[0055] Please see Figure 3 In some implementations, step S02 includes: S022: Calculate the heat generation power based on the charging and discharging current and the internal resistance of the battery cell; S024: Calculate the heat dissipation power based on the cell temperature, ambient temperature, cell heat dissipation coefficient, and heat dissipation area; and S026: Calculate the theoretical temperature rise rate based on the heat generation power, heat dissipation power, and the thermal capacity parameters of the battery cell.

[0056] In S022, the heat generation power Q_gen = I² × R. The internal resistance R can be obtained in several ways. Factory calibration method: During battery pack production, the AC internal resistance (ACIR) or DC internal resistance (DCIR) of the cells is measured at a specific temperature (e.g., 25℃) and a specific SOC (e.g., 50%), and stored as a fixed parameter in the main control module 16. Online identification method: During charging and discharging, R = ΔV / ΔI is calculated using the voltage change ΔV and current change ΔI at current step or pulse moments. For example, when the charging current jumps from 0A to 100A, if the cell voltage rises from 3.2V to 3.25V within 10ms after the current stabilizes, the internal resistance R = (3.25 - 3.2) / 100 = 0.5mΩ. This method can reflect the change in internal resistance with temperature and SOC in real time, but requires precise voltage sampling and timing control. Lookup table method: A two-dimensional array of R as a function of temperature T and SOC is established in advance. In step S022, R is obtained by interpolation based on the current temperature and SOC.

[0057] In S024, the heat dissipation power Q_diss = k × A × (T_cell - T_amb). The product of the heat dissipation coefficient k and the heat dissipation area A, kA, can be calibrated as a single parameter K_diss (unit: W / ℃), thus avoiding the need to measure k and A separately. Calibration method: Place the battery pack in a constant temperature chamber and heat it with a constant power P_heat, recording the steady-state temperature difference ΔT. Then, K_diss = P_heat / ΔT. For example, with 50W heating, when the cell temperature stabilizes at 45℃ and the ambient temperature is 25℃, ΔT = 20℃, then K_diss = 2.5W / ℃.

[0058] In S026, the theoretical temperature rise rate ΔT_theory = (Q_gen - Q_diss) / C × Δt. The selection of Δt should match the actual temperature rise rate calculation window. If the actual temperature rise rate uses a 60-second window, the theoretical temperature rise rate should also be calculated as the cumulative temperature rise over 60 seconds. The Δt in the formula should be the window length (e.g., 60 seconds), not the sampling period. In practice, the theoretical instantaneous temperature rise rate per second can be calculated, and then accumulated over 60 seconds to obtain the theoretical temperature rise over 60 seconds. Alternatively, the average value can be used directly: ΔT_theory = (Q_gen_avg - Q_diss_avg) / C × 60 (unit: °C / min). Q_gen_avg and Q_diss_avg are the average values ​​over 60 seconds.

[0059] The thermal runaway early warning and protection method of this application calculates the heat generation power through internal resistance and current, and the heat dissipation power through temperature difference, heat dissipation coefficient, and heat dissipation area. It then combines these with heat capacity to derive the theoretical temperature rise rate. Thus, the theoretical temperature rise rate reflects the dynamic balance between Joule heating and heat dissipation, allowing the reference value to change in real time with charging and discharging current, ambient temperature, and cell temperature. Compared to the fixed lookup table method, this method better reflects the actual physical process, thereby significantly improving the accuracy of anomaly detection.

[0060] Please see Figure 4 In some embodiments, before step S022, step S02 further includes: S028: Update the internal resistance of the cells based on the cumulative charge and discharge amount or operating time of the battery pack.

[0061] Internal resistance updates can be triggered based on accumulated charge and discharge times. The main control module 16 maintains an accumulator, Ah_sum, to record the total charge and discharge ampere-hours of the battery pack 12 since its first use or the last internal resistance update. When Ah_sum exceeds a preset threshold (e.g., 1000 Ah), an online internal resistance identification is initiated. The identification method is as follows: After the battery pack 12 is in a static state (current < 0.5A) for 30 seconds, a short-duration pulse current (e.g., charging at 50A for 1 second) is applied to the battery pack, while simultaneously recording the cell voltage response at a high sampling rate (e.g., 1kHz). The steady-state voltage V0 before the pulse starts, the current I_pulse during the pulse, and the voltage V1 at the end of the pulse are used to determine the internal resistance R = (V1 - V0) / I_pulse. To avoid polarization effects, an extrapolation method based on the voltage rebound curve after the pulse can be used. Another method is to not apply an active pulse but instead use the naturally occurring current changes during charging and discharging (e.g., changes in the charging pile's step current) for identification. Running time update method: The real-time clock (RTC) of the main control module 16 records the cumulative running time. An internal resistance update is triggered every 30 days, using the current step during normal operation (such as the instantaneous charging start-up) to calculate the internal resistance. The updated internal resistance value is stored in the EEPROM or Flash of the main control module 16 and used for subsequent heat generation power calculations. During the internal resistance update process, it is necessary to ensure the cell temperature remains stable (within ±1℃) to avoid temperature interference with the internal resistance measurement.

[0062] The thermal runaway early warning protection method of this application can update the internal resistance of the battery cell online or offline. Since the internal resistance of the battery cell changes with aging, temperature, and state of charge, a fixed factory-set internal resistance can lead to deviations in the calculation of heat generation power. By updating the internal resistance periodically or as needed, the theoretical temperature rise rate model can adapt to the health status of the battery cell, further reducing false alarms and missed alarms, and extending the effective service life of the thermal runaway early warning protection method.

[0063] Please see Figure 5 In some embodiments, the predetermined threshold includes a first threshold and a second threshold, wherein the second threshold is greater than the first threshold; step S03 includes: S032: When the ratio is greater than or equal to the first threshold but less than the second threshold, control the energy storage system to perform an early warning action; S034: When the ratio is greater than or equal to the second threshold, control the energy storage system to perform a protection action.

[0064] The specific values ​​of the first and second thresholds need to be calibrated based on the battery type, safety level, and application scenario. For example, for lithium iron phosphate batteries, due to their high thermal stability, the first threshold can be set to 1.8 and the second threshold to 2.5; for ternary lithium batteries, with a higher risk of thermal runaway, the first threshold can be set to 1.3 and the second threshold to 1.8. The calibration method is as follows: Select a batch of new battery packs and perform normal charge and discharge tests at different temperatures and rates. Statistically analyze the distribution of the ratio K, and take the 99.5th percentile of K as the first threshold, ensuring that the false alarm rate is below 0.5% under normal operating conditions. Then, perform abuse tests such as overcharging and heating on the same batch of battery packs, record the K value before thermal runaway occurs, and take the minimum K value among all tests as the second threshold, ensuring that protection is triggered before thermal runaway. In addition, the first and second thresholds can also be dynamically adjusted, for example, by appropriately reducing the thresholds according to the degree of battery aging to improve warning sensitivity.

[0065] A residential lithium iron phosphate energy storage system, after calibration, has a first threshold set to 1.6 and a second threshold set to 2.2. During normal charging, the ratio K typically fluctuates between 0.8 and 1.4. One day, due to a micro-short circuit inside the cell, the actual temperature rise rate increases, and K reaches 1.7, triggering a warning action (reducing the charging current and triggering an alarm). After the user reduces the load, K drops to 1.4, and the warning automatically stops. If the user does not intervene, K continues to rise to 2.3, triggering a protection action (cutting off the circuit) to prevent thermal runaway.

[0066] Correspondingly, please refer to Figure 2 In some embodiments, the early warning protection module 18 includes an audible and visual alarm module 182 and a charging and discharging circuit control module 184. The audible and visual alarm module 182 is used to issue audible and visual alarm signals, and the charging and discharging circuit control module 184 is used to control the on / off of the main relay between the battery pack 12 and the external load or charging equipment.

[0067] If a warning action is triggered, the main control module 16 drives the audible and visual alarm module 182 (e.g., a buzzer and LED indicator) to emit an intermittent audible and visual alarm. If a protection action is triggered, the main control module 16 controls the charging and discharging circuit control module 184 (e.g., disconnecting the main relay through an isolation drive circuit) to cut off the charging and discharging circuit of the battery pack 12.

[0068] In addition, the energy storage system 10 may also include a heat dissipation module 1c. The heat dissipation module 1c (such as a fan or liquid-cooled pump) is used to activate in the event of an alarm to enhance heat dissipation. The main control module 16 may also activate the heat dissipation module 1c (e.g., turn on the fan or liquid-cooled pump) when an alarm action is triggered to enhance heat dissipation and suppress temperature rise.

[0069] The thermal runaway early warning protection method of this application sets two threshold levels, which trigger different levels of response. When the ratio is greater than or equal to the first threshold but less than the second threshold, it indicates a minor abnormality in the battery cell. In this case, only a warning action is executed, such as reducing the charging and discharging power or issuing an audible and visual alarm, which alerts the user without forcibly shutting down the device, balancing safety and user experience. When the ratio is greater than or equal to the second threshold, it indicates a serious abnormality in the battery cell. In this case, a protection action is executed, such as cutting off the charging and discharging circuit and forcibly isolating the fault, avoiding the user inconvenience caused by a one-size-fits-all approach, while providing the strongest protection in truly dangerous situations.

[0070] In some implementations, the warning action includes reducing the charging and discharging power of the battery pack and / or issuing an audible and visual alarm signal; the protection action includes disconnecting the charging and discharging circuit of the battery pack.

[0071] Reducing charging and discharging power can be achieved by the main control module 16 sending a power limiting command to the energy storage converter (PCS). The command protocol can be a specific message on the CAN bus (such as ID 0x123, with the data field containing the maximum allowable charging current value). For example, when a warning is triggered, the main control module 16 sends a command via CAN to limit the maximum charging current from 100A to 50A and the maximum discharging power from 10kW to 5kW. The audible and visual alarm signals include: the main control module 16 outputs a PWM waveform (e.g., 1Hz, 50% duty cycle) through a GPIO pin to drive the buzzer BZ1 (not shown) to emit intermittent "beep" sounds; simultaneously, it drives the red LED indicator D1 (not shown) to flash at a frequency of 1Hz through another GPIO. To disconnect the charging and discharging circuit, the main control module 16 outputs a high (or low) level through a GPIO to drive an optocoupler-isolated MOSFET driver (such as TLP250), thereby controlling the coil of the main relay K1 (not shown) to be energized or de-energized. Relays typically use normally open contacts; the contacts open when the coil is de-energized and close when the coil is energized. During protection operation, the main control module 16 de-energizes the coil, opening the contacts and disconnecting the battery pack 12 from external electrical connections. For safety, a pre-charging circuit is usually also required to prevent a large current surge when the relay is closed.

[0072] Warning actions may also include activating the cooling fan to enhance convection cooling. The main control module 16 drives the fan relay via a GPIO or directly controls the fan speed using PWM. Protection actions may also include sending a start signal to the fire suppression system (e.g., dry contact closure) to trigger sprinklers or gas extinguishing systems. For liquid-cooled systems, protection actions may shut down the liquid cooling pump to prevent leakage.

[0073] The thermal runaway early warning and protection method of this application reduces the charging and discharging power when an early warning is issued, which can slow down the deterioration rate of abnormal temperature rise. At the same time, an audible and visual alarm prompts on-site personnel to pay attention to the abnormal state. When protection is issued, the charging and discharging circuit is directly cut off, fundamentally eliminating the energy source of thermal runaway. This method can effectively intervene in the abnormal development without affecting the normal use of the energy storage system due to over-protection.

[0074] Please see Figure 6 In some embodiments, after step S032, step S03 further includes: S036: When the ratio drops below the third threshold, control the energy storage system to stop the warning action; the third threshold is less than the first threshold.

[0075] The third threshold, also known as the recovery threshold, is typically 0.2-0.3 lower than the first threshold. For example, if the first threshold is 1.5, the third threshold can be set to 1.3. When the ratio falls from a high level and remains below 1.3, the main control module 16 executes a stop warning action: it shuts down the buzzer and LED, and sends a power limit recovery command to the PCS via the CAN bus, restoring the charging and discharging current and power to normal values. To prevent false recovery caused by noise, a debounce time is usually set; for example, the stop action is only executed if the ratio is below 1.3 for five consecutive sampling periods (10 seconds per period). Alternatively, the recovery threshold can be set to a fixed hysteresis (e.g., 0.3) lower than the first threshold, and the recovery threshold will automatically follow regardless of how the first threshold is adjusted.

[0076] After the energy storage system 10 triggers the warning action, the user reduces the load, the cell temperature rise rate decreases, and the ratio drops from 1.6 to 1.4. At this point, it is still above the third threshold of 1.3, and the warning continues. When the ratio further drops to 1.2 and remains there for 30 seconds, the main control module 16 determines that the anomaly has been resolved, stops the warning, and resumes normal charging and discharging. If the ratio fluctuates around 1.3 (e.g., 1.31→1.29→1.32), due to the de-jittering time, frequent start-stop cycles will not occur.

[0077] The thermal runaway early warning protection method of this application incorporates an automatic recovery mechanism with hysteresis. When the abnormal temperature rise subsides and the ratio of the actual temperature rise rate to the theoretical temperature rise rate falls back to a third threshold (e.g., 1.3 times) below the first threshold, the early warning action is automatically lifted, allowing the energy storage system to return to normal operation. This avoids alarm jitter caused by frequent triggering and lifting of early warnings near the critical value, and requires no manual intervention, thus improving the level of intelligence and user experience.

[0078] Please see Figure 7 In some embodiments, after S034, step S03 further includes: S038: Locking protection action; and S03a: When an external manual reset signal is received, the protection action is released.

[0079] The locking protection action means that once the protection action (relay disconnection) is executed due to the ratio exceeding the second threshold, the main control module 16 writes the protection status flag to a non-volatile memory (e.g., Flash or EEPROM) and enters a fault latching state in the software. In this state, the main control module 16 will keep the relay disconnected and will not automatically close the circuit even if the ratio subsequently drops to the normal range. The external manual reset signal can come from a physical button (reset button S1, not shown in the figure). When pressed, a falling edge interrupt is generated. After detecting the interrupt, the main control module 16 reads the current cell temperature, voltage, and other parameters to determine whether the safe closing conditions are met (e.g., no voltage abnormality, no over-temperature, no short circuit). If met, the latching flag is cleared and relay control is restored; if not met, the latching state is maintained and an alarm is issued (e.g., a fault code is displayed via LED). Remote reset can receive reset commands from the cloud platform through a communication interface (e.g., a 4G module). After parsing the command, the main control module 16 executes the same safety judgment logic.

[0080] It can be required that an insulation resistance test be performed before manual reset, for example, measuring the insulation resistance between the positive and negative terminals of the battery pack and ground; a value greater than 1MΩ is required for reset. For unattended sites, an automatic reset attempt strategy can be designed: automatically check the safety conditions every 24 hours; if the conditions are met, automatically reset and close the circuit; if three consecutive failures occur, the circuit is permanently latched.

[0081] The thermal runaway early warning and protection method of this application adopts a latching mechanism for protection actions, which requires manual inspection and troubleshooting before manual reset. This prevents the energy storage system from automatically restarting when the fault is not eliminated, avoids repeated risks, complies with the requirement in safety specifications that serious faults must be manually confirmed, and ensures the safe use of the energy storage system.

[0082] Please see Figure 8 In some implementations, step S03 further includes: S03b: Stores cell temperature, charging / discharging current, and ratio in the energy storage module of the energy storage system.

[0083] Correspondingly, the energy storage system 10 includes a storage module 1b. The storage module 1b can be the internal Flash memory of the main control module 16, or an external EEPROM, FRAM, or SD card. Each time an alarm or protection is triggered, the main control module 16 assembles the current timestamp, cell temperature (accuracy 0.1℃), charging / discharging current (accuracy 0.1A), ratio K (accuracy 0.01), and action type (1 alarm / 2 protection) into a record and writes it sequentially to the storage area. Each record occupies 8-16 bytes. Assuming a storage capacity of 32KB, approximately 2000 records can be stored. When the storage space is full, a FIFO (First-In, First-Out) method can be used, overwriting the oldest data. For ease of post-event analysis, only high-frequency waveform data (one point per second) for 5 minutes before and after the abnormal event can be stored. When the data volume is large, compression algorithms can be used, or only feature values ​​can be stored. Data can be exported to the cloud via a local maintenance interface (RS485, Bluetooth) or a remote communication module.

[0084] An energy storage system triggered an alarm at 14:23:45 on March 15, 2025, as recorded below: [2025-03-15 14:23:45, T_cell=42.3℃, I=52.1A, K=1.62, Action=Warning]. Maintenance personnel read this record via a mobile app, determined that the battery cell was abnormal, and arranged for repairs.

[0085] The thermal runaway early warning and protection method of this application stores key data at abnormal moments in the storage module, providing objective evidence for post-fault analysis. This helps to determine the cause of the abnormality (e.g., whether it is caused by increased internal resistance, poor heat dissipation or micro-short circuit), and also facilitates manufacturers to optimize algorithms or users to trace the accident process, thereby improving the maintainability and accident analysis capabilities of the energy storage system.

[0086] Please see Figure 9 In some implementations, step S03 further includes: S03c: The communication module of the control energy storage system sends early warning and protection information to the remote monitoring platform or user terminal.

[0087] Correspondingly, the energy storage system 10 includes a communication module 1a. The communication module 1a can take various forms. For scenarios with broadband networks, a Wi-Fi module can be used to POST data to the cloud server via the HTTP protocol. For outdoor base stations without broadband, a 4G CAT1 module can be used to upload data via the MQTT protocol. For industrial energy storage, CAN bus or RS485 is commonly used to send information to the energy management system (EMS), which then transmits it via Ethernet. Warning and protection information should include: unique device identifier (SN), timestamp, cell temperature, current, ratio, action type, battery pack voltage, etc. The data format can be JSON: {"sn":"ESS1001","time":"2025-03-15T14:23:45Z","temp":42.3,"current":52.1,"ratio":1.62,"action":"warning"}. The user terminal can be a mobile app, which receives push notifications from the cloud platform (such as JPush or APNs) and displays the information in real time. For remote monitoring platforms, historical data query and curve plotting functions should also be supported.

[0088] In scenarios where internet access is unavailable, alarm SMS messages can be sent to maintenance personnel's mobile phones via an SMS gateway. The main control module 16 controls the 4G module to send text messages via AT commands.

[0089] The thermal runaway early warning and protection method of this application pushes early warning and protection information to the cloud or user's mobile phone in real time through a communication module (such as CAN bus or 4G module), so that unattended energy storage systems can also know the battery status in a timely manner, which facilitates rapid response and prevents the accident from escalating. It is especially suitable for distributed energy storage, residential energy storage and other scenarios, and improves the remote management capability of energy storage systems.

[0090] In some embodiments, the energy storage system 10 includes multiple cell temperature probes for detecting the temperature of multiple cells at multiple locations within the battery pack. S01 includes: S012: Obtain the temperature of multiple battery cells; S02 includes: S02a: Calculate multiple theoretical temperature rise rates corresponding to multiple cell temperatures; S03 includes: S03d: Calculates multiple ratios between actual temperature rise rates and corresponding theoretical temperature rise rates for multiple cell temperatures; and S03e: If any ratio exceeds a predetermined threshold, control the energy storage system to perform an early warning protection action.

[0091] In other words, the acquisition module 14, or more precisely, the cell temperature acquisition module 142, includes multiple cell temperature probes. The arrangement principle for these probes is that each cell module in the battery pack 12 has at least one probe, with a focus on monitoring areas prone to heat generation (such as the center of the battery pack, near the positive and negative electrode connections, and locations with poor heat dissipation). The number of probes can be 4, 8, 16, etc., depending on the number of cells and the heat distribution. Each probe is independently connected to the analog input channel of the acquisition module 14, and the main control module 16 can read the temperature value of each channel. For each probe location, the theoretical temperature rise rate and the actual temperature rise rate at that location are calculated, yielding their respective ratios, K_i. The judgment logic is: if any i exists such that K_i ≥ a first threshold, an early warning is triggered; if any i exists such that K_i ≥ a second threshold, protection is triggered. This allows for the detection of local anomalies, such as a cell having a high internal resistance due to a manufacturing defect, resulting in a significantly higher temperature rise at that location compared to other locations. In addition, it can compare the temperature difference or ratio difference between different probes. When the difference exceeds the set value (such as the maximum temperature difference > 10℃), it is also judged as abnormal.

[0092] A 16-cell battery pack has one probe positioned between cells #8 and #9, and another probe positioned between cells #1 and #2. During charging, the contact resistance at positions #8 and #9 increased due to loose connecting tabs, resulting in a temperature rise rate of 5°C / min, while the temperature rise rate at other positions was only 1°C / min. The theoretical temperature rise rate was 1.2°C / min, with a ratio K=4.17, far exceeding the second threshold of 2.0. The system immediately triggered protection, cutting off the circuit and preventing the connecting tabs from overheating and catching fire.

[0093] A fiber optic grating temperature sensing system can be used, integrating multiple gratings on a single optical fiber. This system can simultaneously measure the temperature at dozens of points within the battery pack and is unaffected by electromagnetic interference. The main control module 16 reads the wavelength changes of each grating using a demodulator and converts them into temperature values.

[0094] The thermal runaway early warning method of this application sets cell temperature probes at multiple locations within the battery pack to acquire multiple cell temperatures and calculates the theoretical and actual temperature rise rates for each location, obtaining multiple ratios. When any ratio exceeds a predetermined threshold, the method immediately triggers an early warning. Since temperatures may differ at different locations within the battery pack (e.g., localized cell aging, micro-short circuits, or uneven heat dissipation), monitoring only a single temperature point may not promptly detect localized anomalies. By calculating the theoretical temperature rise rate for multiple locations and comparing the actual temperature rise rate, the thermal runaway early warning method can more sensitively identify abnormal temperature rises at any location within the battery pack, preventing the spread of thermal runaway due to undetected localized overheating. Simultaneously, the theoretical temperature rise rate for each location is dynamically calculated based on the current heat generation and dissipation conditions at that location, adapting to different operating conditions and effectively avoiding missed or false alarms caused by fixed thresholds, further improving the safety and reliability of the energy storage system.

[0095] In this specification, the terms "specifically," "furthermore," "particularly," "understandably," etc., refer to specific features, structures, materials, or characteristics described in connection with embodiments or examples that are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0096] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of executable request code comprising one or more steps for implementing a particular logical function or process, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order according to the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0097] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A thermal runaway early warning and protection method for an energy storage system, characterized in that, The method includes: The cell temperature, charging and discharging current, and ambient temperature of the battery pack of the energy storage system are obtained. Based on the cell temperature, the charging / discharging current, the ambient temperature, and the heat generation, heat dissipation, and heat capacity parameters of the cells in the battery pack, the theoretical temperature rise rate of the cell is calculated; and If the ratio of the actual temperature rise rate of the battery cell to the theoretical temperature rise rate exceeds a predetermined threshold, the energy storage system is controlled to perform an early warning protection action.

2. The thermal runaway early warning and protection method according to claim 1, characterized in that, The calculation of the theoretical temperature rise rate of the battery cell includes: Calculate the heat generation power based on the charging and discharging current and the internal resistance of the battery cell; Calculate the heat dissipation power based on the cell temperature, the ambient temperature, the cell's heat dissipation coefficient, and the heat dissipation area; and The theoretical temperature rise rate is calculated based on the heat generation power, the heat dissipation power, and the thermal capacity parameters of the battery cell.

3. The thermal runaway early warning and protection method according to claim 2, characterized in that, Before calculating the heat generation power, the calculation of the theoretical temperature rise rate of the battery cell further includes: The internal resistance of the battery cell is updated based on the cumulative charge / discharge amount or operating time of the battery pack.

4. The thermal runaway early warning and protection method according to claim 1, characterized in that, The predetermined threshold includes a first threshold and a second threshold, wherein the second threshold is greater than the first threshold; The control of the energy storage system to perform early warning and protection actions includes: When the ratio is greater than or equal to the first threshold but less than the second threshold, the energy storage system is controlled to perform an early warning action. When the ratio is greater than or equal to the second threshold, the energy storage system is controlled to perform a protection action.

5. The thermal runaway early warning and protection method according to claim 4, characterized in that, The warning action includes reducing the charging and discharging power of the battery pack and / or issuing an audible and visual alarm signal; the protection action includes disconnecting the charging and discharging circuit of the battery pack.

6. The thermal runaway early warning and protection method according to claim 4, characterized in that, After the energy storage system is controlled to perform an early warning action, the method of controlling the energy storage system to perform an early warning protection action further includes: When the ratio drops below a third threshold, the energy storage system is controlled to stop the warning action; the third threshold is less than the first threshold.

7. The thermal runaway early warning and protection method according to claim 4, characterized in that, After the energy storage system is controlled to perform a protection action, the step of controlling the energy storage system to perform an early warning protection action further includes: Locking the protection action; and The protection action is released upon receiving an external manual reset signal.

8. The thermal runaway early warning and protection method according to claim 1, characterized in that, The method of controlling the energy storage system to perform early warning and protection actions also includes: The cell temperature, the charging and discharging current, and the ratio are stored in the storage module of the energy storage system.

9. The thermal runaway early warning and protection method according to claim 1, characterized in that, The method of controlling the energy storage system to perform early warning and protection actions also includes: The communication module of the energy storage system is controlled to send early warning and protection information to the remote monitoring platform or user terminal.

10. The thermal runaway early warning and protection method according to claim 1, characterized in that, The energy storage system includes multiple cell temperature probes, which are used to detect the temperature of multiple cells at multiple locations within the battery pack. The acquisition of the cell temperature, charging / discharging current, and ambient temperature of the battery pack of the energy storage system includes: Obtain the temperature of the multiple battery cells; The calculation of the theoretical temperature rise rate of the battery cell includes: Calculate the theoretical temperature rise rates corresponding to the temperatures of the multiple battery cells; The control of the energy storage system to perform early warning and protection actions includes: Calculate multiple ratios of the actual temperature rise rate to the theoretical temperature rise rate corresponding to multiple cell temperatures; and If any of the ratios exceeds the predetermined threshold, the energy storage system is controlled to perform the early warning protection action.

11. An energy storage system, characterized in that, include: Battery pack; The data acquisition module is used to collect the cell temperature, charging and discharging current and ambient temperature of the battery pack. and The main control module is used to execute the thermal runaway early warning and protection method according to any one of claims 1 to 10; The early warning and protection module is used to execute the aforementioned early warning and protection actions.

Images